Citation
Pattern mapping in plane motion analysis

Material Information

Title:
Pattern mapping in plane motion analysis
Creator:
Fail, R. Wallace, 1954-
Publication Date:
Language:
English
Physical Description:
xii, 104 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Centroids ( jstor )
Coordinate systems ( jstor )
Digital images ( jstor )
Image analysis ( jstor )
Image processing ( jstor )
Image rotation ( jstor )
Kinetics ( jstor )
Pixels ( jstor )
Supernova remnants ( jstor )
Tensors ( jstor )
Materials -- Testing ( lcsh )
Motion ( lcsh )
Strains and stresses ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1987.
Bibliography:
Includes bibliographical references (leaves 99-103).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by R. Wallace Fail.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
021380007 ( ALEPH )
AFA8353 ( NOTIS )
18035304 ( OCLC )

Downloads

This item has the following downloads:


Full Text


















PATTERN MAPPING IN
PLANE MOTION ANALYSIS







BY







R. WALLACE FAIL


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1987
































Copyright 1987

by

R. Wallace Fail
































To the Glory of God


Digitized by the Internet Archive
in 2010 with funding from
University of Florida, George A. Smathers Libraries with support from Lyrasis and the Sloan Foundation


http://www.archive.org/details/patternmappingin00fail

















ACKNOWLEDGMENTS


I would like to express my appreciation to the members of my

supervisory committee, Dr. S. S. Ballard, Dr. P. Hajela, Dr. U. H.

Kurzweg, Dr. E. K. Walsh, and Dr. C. E. Taylor, chairman.

Support of the graduate program by former department head, Dr.

K. Millsaps, and current department head, Dr. M. A. Eisenberg, is

gratefully acknowledged.

Special thanks go to Mr. Steve McNeil and Mr. Marc Paquette at

the University of South Carolina for their help and extending the use

of their laboratory facilities to digitize the images used in this

study.

I am indebted to my wife, Robyn, for her unwavering support and

love.
















TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS................................................. iv

LIST OF TABLES.................................................. vii

LIST OF FIGURES.................................................. .ix

ABSTRACT........................................................... xi

CHAPTERS

I INTRODUCTION............................................. 1

I.1 Background........................................ 1
1.2 Purpose of Present Work.............................. 3
1.3 Scope of Present Work................................ 3
1.4 Survey of Previous Work.............................5
1.4.1 Solid Mechanics .............................5
1.4.2 Image Processing/Pattern Recognition.........7

II THEORETICAL FOUNDATIONS...................................8

[I.1 Solid Mechanics.....................................
II.2 Image Processing and Pattern Recognition............14
11.2.1 Digital Image.............................. 14
11.2.2 Preprocessing..............................18
11.2.3 Segmentation ...............................18
11.2.4 Image Features.............................. 24
11.2.5 Feature Extraction.........................26
11.2.6 Known Patterns.............................39
11.2.7 Syntactic Pattern Recognition..............41
11.2.8 Approximation of Syntactic Pattern
Mapping.................................... 42

III IMAGE GENERATION AND ANALYSIS ............................58

II1.1 Synthetic Images...................................58
III.2 Image Analysis.................................. ..60











IV EXPERIMENTS.............................................. 78

IV.1 Test Specimen......................................78
IV.2 Test Equipment..................................... 78
IV.3 Experimental Procedure..............................82

V ANALYSIS OF RESULTS........................................84

V.1 Numerical.......................................... 84
V.2 Experimental............................................. 89

VI CONCLUSIONS AND RECOMMENDATIONS...........................94

VI.1 Conclusions........................................94
VI.2 Recommendations...................................95

APPENDIX BORDER FOLLOWING ALGORITHM..............................98

REFERENCES.........................................................99

BIOGRAPHICAL SKETCH...............................................104





































vi
















LIST OF TABLES


Table Page

1 Unordered spot coordinates in the undeformed image..........45

2 Unordered spot coordinates in the deformed image.............46

3 Ordered spot coordinates in the undeformed image............51

4 Ordered spot coordinates in the deformed image..............52

5 Spot distance from origin in undeformed image...............53

6 Spot distance from origin in deformed image.................54

7 Spot distance from Icon in undeformed image.................55

8 Spot distance from Icon in deformed image...................56

9 Input file for synthetic image generator....................59

10 Image analysis options file.................................60

11 Displacement of each spot...................................61

12 Displacement gradients......................................70

13 Lagrangian strain from displacement gradients...............71

14 Strain from Taylor series...................................72

15 Deformation gradient...................................... 73

16 Green deformation tensor...................................74

17 Right stretch tensor........................................75

18 Rotation tensor......................................... .. 76

19 Lagrangian strain from deformation gradient.................77

20 Summary of strain analysis for synthetic images.............86

21 The effect of SNR on translation and strain calculations....87











22 The effect of gray level difference on translation and
strain calculations............................. ........... 87

23 The effect of spot radius on translation and strain
calculations............................................. 88

24 Images used for rigid-body motion experiments...............89

25 Results of rigid-body rotation experiments...................90

26 Results of rigid-body translation experiments...............92


viii















LIST OF FIGURES


Figure Page

1 The motion of body B through space.......................... 9

2 The motion of the neighborhood of particle p...............10

3 A typical PC-based image processing system.................15

4 A spot and its digital image................................16

5 A pixel and its eight-neighbors............................17

6 The bi-modal histogram of figure 4 .........................19

7 Black and white regions in a bi-modal histogram............21

8 Frame of spot in figure 4.................................. 24

9 Image artifacts........................................... 25

10 One-dimension model.......................................30

11 The discretization of KT and B ........................ ...33

12 Icon rotation measured from the vertical...................35

13 Icon rotation is determined to within a constant by
equation (40) ............................................. 37

14 Patterns that are easily recognized and analyzed............40

15 Patterns of Figure 14 with Icons............................40

16 A pattern and its context-sensitive language................42

17 Image addresses and offsets.................................43

18 A simple example of motion................................. 44

19 The deformed image and its reference position...............48

20 Beam fabricated from CR-39.................................79










21 Spot pattern on beam.......................................79

22 Experimental set-up of load frame.........................80

23 Set-up for four-point bending..............................80

24 Set-up for cantilever beam.................................81

25 Rigid-body motion set-up..................................81
















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PATTERN MAPPING IN
PLANE MOTION ANALYSIS

BY

R. WALLACE FAIL

May 1987


Chairman: C. E. Taylor
Major Department: Engineering Sciences

A new, highly automated method for measuring plane motion,

pattern mapping, has been developed for rigid-body motion and strain

analysis. Pattern mapping employed image processing and syntactic

pattern recognition principles to recognize a known pattern before

and after motion. Using the Lagrangian definition of motion, points

in the two images were mapped, and the map was used to determine

rigid-body motion and strain. Whole-image analysis was thoroughly

demonstrated.

Known patterns were, in this case, rectangular grids of photo-

dots. These patterns were applied to the specimen by contact print-

ing. An icon was included in the pattern, thereby accommodating

rigid-body rotations up to one revolution.

Motion functions were selected by the user and easily changed to

suit the particular application. Function coefficients were deter-

mined by least squares.










Accuracy depended primarily on the signal-to-noise ratio and the

gray level difference between "black" and "white." An increase in

these variables improved accuracy.

User and computer time requirements were quite modest. Images

were processed with little or no user interaction, and output speci-

fications were automatically read from a disk file. Depending on

output specifications, computer time (VAX 11/750) for an 11X11

pattern in a 384X374 image was 25 to 38 CPU seconds.
















CHAPTER I
INTRODUCTION



I.1 Background

The primary purpose in experimental solid mechanics is to deter-

mine stress and motion (strain, rigid-body translation, and rotation)

from measurements of physical quantities. For example, stress may be

computed from measurements of load and cross-sectional area, and

strain is measured by changes in characteristic lengths. Many

methods (mechanical, electrical, optical, etc.) are available to

analyze stress and motion [1-4].

Optical methods play an important role in experimental mechan-

ics. They have a number of advantages which include [5,6]

1. Noncontacting measurements do not interfere with material

response.

2. Full-field analysis is usually possible.

3. Highly accurate results are usually possible.

4. Response times are as fast as light, thereby accommodating

dynamic investigations.

5. Actual structures under live loads can frequently be ana-

lyzed.

6. Effects of actual boundary conditions can be studied in

detail.










Although optical methods offer many advantages, they do have

several undesirable features. For example, some methods are sensi-

tive to small extraneous vibrations and require expensive, delicate

equipment. Perhaps the most undesirable feature is the fact that

most of these methods do not directly yield their wealth of informa-

tion about stress or motion in a readily usable form.

The extraction of this information is time-consuming and re-

quires special skills. An expert may spend hours reducing data and,

all too often, the amount of data analyzed is limited simply because

automated analysis is not available. Furthermore, results can be

affected by the investigator's particular techniques.

Digital computers are opening doors to automation, increased

accuracy of existing methods, and previously intractable or unknown

techniques. Some of the latest innovations employ digital image

processing and pattern recognition (IP/PR) technologies which utilize

digital computers and video imaging equipment for direct data acqui-

sition.

The fields of IP/PR are rapidly expanding and are well-

established in many areas. In experimental mechanics, their poten-

tial for efficiently gathering vast quantities of data consistently

and accurately is being quickly developed. These technologies are

freeing experimentalists from hours of painstaking labor and provid-

ing them with a new avenue to existing numerical methods and analyti-

cal tools [7-14].










1.2 Purpose of Present Work

The IP/PR technologies have their price. For example, IP/PR

programs that use fast Fourier transforms or correlation (template

matching) are frequently considered computationally expensive because

they entail large numbers of (relatively slow) floating-point calcu-

lations. Interactive programs provide great flexibility, but they

need the user's constant attention. Iterative and backtracking

(systematically retracing previous steps) methods can require sub-

stantial amounts of computer memory for intermediate data storage

[15-18].

Significantly reducing these time and memory requirements is

certainly a desirable goal. Perhaps this is possible by rethinking

the problem and developing a highly automated method which capital-

izes on the strengths of the digital computer. Thus, the purpose of

this work is to develop such a method, pattern mapping, to obtain

data via digital imaging equipment and subsequently determine plane

motion.



1.3 Scope of Present Work

Although it was not apparent in the beginning, the concepts

developed in this investigation are readily applied to the photodot

(grid) method [19]. The scope includes this photodot adaptation.

Actually, this investigation encompasses a variety of scientific

disciplines. Perhaps the best approach is to consider the attributes

of each and work toward a goal that is obtainable in a reasonable

amount of time. Specific objectives included in the scope are to










1. guarantee unique mapping

2. provide subpixel registration of 0.1 pixel

3. provide a wide range of motion measurement

4. provide whole-image analysis

5. reduce computer and user time as much as possible

6. provide highly automated analysis.

(Pixels are discrete elements in a digital image, and whole-image

implies results are computed for every photodot.) Special emphasis

is on accuracy and increased computation efficiency.

Several laboratory experiments using a CR-39 (plastic) prismatic

beam were included in the study. The experiments were translation,

rotation, general rigid-body motion, four-point bending, and

cantilever beam. All experiments were static cases of motion in a

plane normal to the optical axis. A rectangular (11X11) pattern of

photodots was contact printed on the lateral side of the beam.

For convenience, motion is defined only in terms of the

Lagrangian formulation. Evaluation of motion functions are limited

to a linear and a second order approximation. The least squares

method of curve fitting is used exclusively, and normal equations are

solved by matrix inversion using Gauss-Jordan elimination.

Synthetic images are used to debug analysis programs and inves-

tigate the influence of various factors. Motion in the images in-

clude affine transformations and rigid-body rotation about the image

center. Interpolation of between-pixel intensities is bilinear, and

system-induced distortions are assumed eliminated during preprocess-

ing. Noise has a zero mean and selectable Gaussian or uniform





5



distributions. Image size is 256X256 pixels which are unsigned, 16

bits or less. The camera position is assumed fixed.

Segmentation and feature extraction (dividing the image into

subregions and obtaining useful information) include neither prefil-

tering nor histogramming (gray levels) and are based on the random

selection of a few pixels. Processing is automatic and direct

because known patterns in the image are always completely visible and

recognizable. Feature space includes gray levels, areas, centroids,

and second moments. The pattern language consists of only one sen-

tence, and it is context-sensitive.



1.4 Survey of Previous Work

1.4.1 Solid Mechanics

Early applications of image processing in experimental mechanics

involved the analyses of fringes. The fringes were from photoelasti-

city, moire, or holographic interferometry. Burger [7] and Chen [8]

describe several of these applications in detail.

Another image processing technique applied to mechanics is cor-

relation. Researchers [9,20-24] determine plane displacements (ui,

i=1,2) and their gradients (3ui/aXj, i,j=1,2) by comparing digitally

recorded images of the same random pattern. For example, a pattern

of black speckles (e.g., minute specks of black paint) on a white

field is recorded as integer gray levels at discrete points

(pixels). Two images are used in the analysis: one before and one

after motion. By comparing (spatially registering) small subimages

from each, displacements and their gradients are determined at










selected points in the (original) pattern. Sigler and Haworth [25]

combine holography with correlation to determine motion. Thus, the

correlation method has the advantages of using coherent or

conventional light sources and easily applied patterns.

The grid method is one of the oldest and simplest methods used

in the analysis of strain. One simply applies reference marks

(lines, spots, scratches, etc.) to the surface of a specimen and

records the original distances between marks. Next, the specimen is

deformed and the distances are again recorded. Then, changes in the

distances are determined and strain is computed [1,2,26-32].

Although simple, the grid method is tedious and traditionally

limited to larger strains. Typical strain measurements are greater

than 0.001 inches-per-inch; the results depend on the care exercised

and the instrumentation [19,26,28]. Even when optical comparators

are used for the smaller strains, many man-hours are frequently

needed to complete the analysis of large grids.

Examining a specimen through a comparator and determining the

position of a grid mark is also a subjective process. Some investi-

gators define a mark's position by its left edge, some by its right

edge, and some by a variety of definitions. All these definitions

are subject to error caused by rotation, the nonlinear characteris-

tics of the recording medium (if photographically recorded), grid

application techniques, lighting, operator fatigue, etc. [19].

Parks [32] suggested computer technology might improve the

accuracy of the method and would certainly reduce the labor










involved. Previous efforts to fully computerize the grid method have

not been reported in the literature.

Sevenhuijesen [33,34] reported some feasibility studies with a

Reticon photo-diode array camera. Using an eight-bit digitizer and

digitizing the signal from each diode, he obtained a resolution of a

few hundred micro inches-per-inch. Images were recorded photographi-

cally without the aid of an image processing system.

1.4.2 Image Processing/Pattern Recognition

Several researchers active in IP/PR have been interested in two-

and three-dimensional motion analysis. Most efforts have concentrat-

ed on the motion of rigid bodies, clouds, the human body, optical

flow, etc. Motion parameters were obtained by a variety of iterative

processes [35-38].

Huang and Tsai [39-43] proposed a direct method of mapping the

three-dimensional motion of a plane using two, time-sequential

images. Mapping parameters were determined uniquely by solving a set

of linear equations. Unfortunately, motion parallel to the optical

axis could not be determined, and "small" approximations were made

for strain and rotation. Mathematical proofs guaranteeing unique

mapping between images were included.
















CHAPTER II
THEORETICAL FOUNDATIONS



II.1 Solid Mechanics

The description of motion used here is limited to the Lagrangian

description and is consistent with the classical field theories of

modern continuum mechanics [44-48]. The Lagrangian description is

sufficient to determine the directions of the principal axes of

strain and the magnitudes of the principal stretches. It is, there-

fore, a fully general measure of strain and is capable of describing

small or large strains. For convenience, all reference frames are

Cartesian in three-dimensional Euclidean space.

In Figure 1, body $ consists of a set of particles which assume

a continuous progression of configurations in time. The undeformed

configuration is chosen as the reference state where particle P

occupies the "material" position X at time t=to. Subsequent motion

carries particle P to the "spatial" position x in the deformed or

"after motion" state at time t. Although each state may be defined

in terms of its own coordinate system, material and spatial coordi-

nates are usually measured with respect to the same coordinate

axis. The motion of a point is described by the relation of the

coordinates between the two states. Mathematically, motion is ex-

pressed by a one-to-one mapping x symbolized as









x = x(X,t)


or x. = xi(X IX2,X3 t)
2. 1 2 3'


Figure 1. The motion of body B through space.


The deformation

derivative or gradient


gradient of the motion x is defined as the

of x at X and is denoted by


3x
F = or
- aX


3x.
F. .
ii DXj


where F is the fundamental quantity for the analysis of local proper-

ties of deformation, and the physical components are, of course,

dimensionless as are all measures of relative configuration.





10



Alternatively, motion written in terms of the displacement

vector u is


x = X + u(X,t)


or xi = X. + ui(XI,X 2X ,t)
2. i 1 2' 3


And the derivative of u, the displacement gradient, is


au
J =
X


or J.
ij


1
SX.


If two particles in the reference configuration are an infini-

tesimal distance dX apart, then these two particles are an infinites-

imal distance dx apart in the deformed configuration (see Figure

2). Tensor F maps the neighborhood of particle P in the


Figure 2. The motion of the neighborhood of particle P.










reference configuration to its neighborhood in the deformed configur-

ation. Thus, the linear transformation is


dx = F-dX or dx. = F..dX.
S- 1 13 j


Unique mapping is guaranteed as long as the Jacobian J = IFI is

greater than zero and finite.

Since vector dX has length dS and vector dx has length ds,

Lagrangian strain E may be defined in terms of these lengths squared

as


(ds) (dS)2 = 2dX-E.dX


The Green deformation tensor C, referred to the undeformed con-

figuration, gives the new squared length (ds)2 of the element into

which the given element dX is deformed. Symbolically, the relation-

ship is


(ds)2 = dX dX
(ds) = dX-C-dX


where


C = FT.F


-I
Conversely, the Cauchy deformation tensor B gives the initial

squared length (dS) of an element dx in the deformed









configuration. This relationship is written as


(dS)2 = dx.B *dx


where


-1 (F T F-
B : (F ) -F


Lagrangian strain written in terms of the Green deformation

tensor is


E = [C-1]
2 -2


(11)


In terms of the displacement gradient, E is


I ui au 3u auk au
Eij = 2[--X + i + -X]
ij 2 jX. 2X. 3X. jX.
3 3- 1. 3


Strain, computed by a Taylor series [1], is


E au 8u)2 av 2 ,w 2
S11 V 3X 9X aX X


/ Byv 2 au 2 aw 2'
E = A1+2- +(-) +(-) +() -I
22 1 Y a Y aY aY


(12)


(13)


au yv au u u v Dv Dw aw
+-- + + +
1 r Y x xs a 3X aY 3X aY
arcsin[(1 )(
2 (1+E )(1+E22)
11 22


(10)


12 = 21





13



For many engineering applications, the so-called "small dis-

placement" theory is adequate and much easier to implement. By this

theory, strain is



au
11 ax


1 au av
12 2 Y ax
(14)


C21 = E12


av
22 3X



If the Jacobian J is greater than zero and finite, the deforma-

tion gradient written in terms of a rotation tensor R and the right

stretch tensor U, a polar decomposition, is



F = R.U (15)



Tensors R and U are unique and represent rigid-body rotation and

stretch, respectively. Rotation tensor R is orthogonal and produces

a rigid-body rotation between the principal axes of C at X and the
-1
principal axes of B at x. The necessary and sufficient condition

for no local rigid-body rotation (sometimes called "pure strain")

during motion is R=1. Symmetric, positive-definite U produces the

changes in vector lengths during motion and produces rotation, in

addition to R, of all vectors except those in the principal











directions of U. Note that U=1 is necessary and sufficient to define

locally "pure" rigid-body motion (neglecting the trivial case when no

motion occurs), and translation does not change a vector with respect

to the common reference axis. Tensors U and R, written in terms of

F and C, are


U = C1



R = F.U_1
R --F.U


(16)



(17)


As a final note, the deformation at any point may result from

translation, a rigid-body rotation of the principal axes of strain,

and stretches along these axes. These motions may occur in any

order, but their tensorial measures may not. Mathematically,

Lagrangian motion is assumed as a successive application of

1. stretch by U

2. rigid-body rotation by R

3. translation to x.


11.2 Image Processing and Pattern Recognition

II.2.1 Digital Images

A typical image processing system is shown in Figure 3. Conver-

sion of the intensity distribution of a camera image to a form suit-

able for computers, digitizing, is accomplished by the digitizer.

After digitization, the image is transferred to the










computer for further processing. How an image is processed depends,

of course, on the hardware configuration [15,18,35,49-51].




















CCD Camera F ~

Object Microcomputer Digitizer
with Internal Monitor
Digitizer









Figure 3. A typical PC-based image processing system.






A digital image is an approximation of a real, continuous

image. For example, the real image of the spot in Figure 4a is re-

presented in digital form as the matrix of integer numbers in Figure

4b. Each matrix element (or pixel which is dimensionless) corre-

sponds to a discrete point in the real image, and the value of the

element (gray level) is the approximation of the average local illum-

ination intensity. Notice that black has lower gray levels than












white and that the edge region (boundary) between the two has inter-


mediate values. The variation of gray levels in regions of constant


illumination is the effect of noise. Gray levels range from zero to


some maximum which depends on the hardware (e.g., camera, digitizer).















149 151 155 146 163 159 159 N 15 15 162 151 142 145 146 156 145 166 148 165
157 145 156 142 146 150 127 148 158 166 151 153 142 142 157 149 153 148 143
150 155 156 145 146 150 159 166 56152 54 153 155 132 148 144 151 169 164
144 165 156 138 15 146 148 129 1~ 98 115 127 150 15 137 149 152 151 153
15 151 147 156 148 138 83 47 36 51 56 60 3# 142 154 147 163 146 15
146 152 146 134 138 66 45 52 48 49 47 51 74 52 139 152 154 15 146
145 154 14 141 94 56 56 47 50 43 51 46 53 59 91 151 153 153 151
146 159 143 138 53 61 51 50 55 43 48 51 45 56 44 135 154 146 154
150 150 150 10 51 48 36 44 55 49 51 58 53 50 49 105 141 144 15
153 139 152 96 50 50 44 52 48 56 57 50 54 46 47 1t1 15 143 143
156 155 155 18 40 43 46 49 59 41 48 49 55 51 42 193 146 158 152
153 148 166 137 47 51 47 41 48 37 51 50 50 47 33 137 156 153 143
147 148 151 153 95 45 5 55 49 50 51 48 44 43 82 165 147 153 160
147 135 155 159 138 76 55 45 57 53 54 47 63 132 151 158 147 148
145 148 151 162 159 144 82 50 55 54 53 39 70 141 148 152 148 159 152
151 153 156 153 148 132 148 132 149 96 99 143 154 150 145 154 163 159 151
144 141 149 157 148 156 145 158 145 154 142 139 145 145 147 154 161 148 149
157 150 141 144 158 149 166 150 151 142 142 158 145 156 150 150 151 147 149
156 158 146 160 148 140 148 149 151 144 149 149 150 15 153 152 149 157 160
155 147 150 131 150 150 163 150 158 153 147 146 163 136 151 154 153 152 153
151 16 146 148 144 152 153 148 16Q 160 136 157 144 144 14 162 146 148 154


Real Image Digital Image
a) b)





Figure 4. A spot and its digital image. The values, I(x,y), in the
digital image correspond to illumination intensity in the
real image.






Noise corrupts the image gray levels, and the sources may be


classified into two groups: optical and electronic. Two examples of


optical noise are particulate matter in the air and laser speckle.


Electronic noise comes from the discretization process, thermal










sources in the hardware, interference from extraneous electromagnetic

sources, etc. Noise is usually assumed statistically random and

normally distributed with a zero mean, i, and a standard devia-

tion, a Unfortunately, no known method completely eliminates the

effects of noise.

Each array element has neighbors which are commonly called the

eight-neighbors [52]. In Figure 5, the center element, pixel (x,y),

is the reference element or element of interest. The element

directly above it is arbitrarily labeled neighbor 0, and the other

elements are named in a counterclockwise direction. Although ele-

ments along the image's outer array edge do not have all


Figure 5. A pixel and its eight-neighbors.


1 0 7




2 Pixel 6
(x.y)



3 4 5











8 neighbors, the neighbors they do have are also labeled according to

Figure 5.

Spatial and illumination sampling must be sufficient to repre-

sent adequately the real image. Too low a spatial sampling frequency

averages intensity over too large an area (see Figure 2.7 in refer-

ence 15), while too high a sampling frequency results in the loss of

information outside the field of view. In general, increasing inten-

sity corresponds to increased gray levels, but intensities beyond the

capability of the equipment are limited to the maximum gray level.

An increase in the range of gray levels increases image detail (see

Figure 2.8 in reference 15).

11.2.2 Preprocessing

After digitization, the digital image is corrected for geometric

and radiometric distortion. Distortion, introduced by the processing

system, is removed by using the appropriate calibration procedure

[49,53]. This type of correction is frequently called preprocessing

and images in this study are assumed corrected unless otherwise

stated.

II.2.3 Segmentation

The next step in image processing, segmentation, consists of

identifying regions and boundaries of interest in the image

[18,54]. Of course, the difficulty in segmenting an image depends on

its complexity. Since the images developed in this study are rela-

tively simple, they are easily segmented.

Histogramming. Segmentation usually begins with histogram-

ming. When the frequency distribution of the gray levels in an image






19



is computed for the entire image, contrast and the presence of multi-

ple modes are readily observed. Figure 6, the bi-modal histogram of

Figure 4, has an average black of 50 and an average white of 150.

Note, the histogram yields no direct information about the location

of the gray levels in the image [55].









50 -

Contrast
Range
40


Pixel 30
Count


20



10-


0 qLA A. .I
0 20 40 60 80 100 120 140 160 180 200

Gray Levels




Figure 6. The bi-modal histogram of Figure 4.






Usually, computing the histogram for the entire image is not

considered computationally expensive. This is not true in the

present example because the percent of the total computer time

devoted to histogramming the entire image is significant. Since the

images used here are designed to have histograms similar to Figure 6,










the needed information is easily estimated without actually computing

a histogram.

Thresholding. Thresholding follows histograimming and includes

the task of determining the gray level which, in this case, reason-

ably separates regions of black from white. This is frequently

accomplished interactively or by use of a priori information, and the

choice of the optimum threshold is a subject of considerable interest

[56-62].

The alternative approach in this study requires neither inter-

active thresholding nor an optimum threshold. Instead, a good esti-

mate of gray levels that are "definitely black" and an estimate of

the average white level are used. The estimates are derived from

histogram statistics and based on the following assumption: the

histogram is strongly bi-modal. In Figure 6, gray level 53 is

assumed black and the average white gray level is about 151. The

task of estimating these levels is a four-step process.

First, pixels are randomly selected and their average gray level

IA is computed along with the standard deviation oA. A sample size

of 30 or more is statistically sufficient to estimate the mean and

standard deviation of the entire histogram [63,64]. Obviously,

darker pixels are to the left of the average and lighter pixels are

to the right (see Figure 7).

In the second step, the standard deviation is subtracted from

the average, resulting in IB. Algebraically, this is


I = IA A


(18)





21










50


40


30 I

Pixel 1 B A
Count
20






0 20 40 60 80 100 120 140 160 180 200

Gray Levels







Figure 7. Black and white regions in a bi-modal histogram (IA=132,
I=98, I1=53, 12=151, It=102).





Another set of 30 or more pixels whose values are below IB are ran-

domly selected and used to compute the average I1. Thus, the average

black level is approximated by I1.

Chebychev's theorem [63] states that, if a probability distribu-

tion has a mean p and a standard deviation a, the probability of

obtaining a value x which deviates from the mean by at least K stand-

ard deviations is at most Mathematically, this is expressed as
K2
KK


P(|x-U| > Ko) < -- (19)
K













Since only one standard deviation is subtracted from the average,

finding pixels darker than Ig is highly probable. A skew of the

histogram to the left further improves the odds of finding pixels

less than Ig.

The third step is the initial estimation of the average white

gray level, I2. A sufficient random sample whose values are greater

than IA is used to compute the initial estimate of 12.

Finally, the threshold value It is computed using the estimates

for the black and white levels. Symbolically,


1 + 12
I = 2 (20)
t 2



Images designed for this study are assumed to have two

regions: black and white. In a sense, they are binary (region)

images. Thus, segmentation is easily accomplished by locating and

isolating the black regions of interest from the white background.

Raster scanning and border following. Beginning at the origin,

the upper left corner, the image is raster scanned (i.e., along a row

or column) until a black pixel is found. If at least one of the

neighbors is black, a possible black region is assumed--spurious

noise is assumed if the pixel has no black neighbors. If a black

neighbor is found, the border between black and white is followed

around the black region thereby isolating it from the neighboring

white.











Rosenfeld [18] has developed a general-purpose border-following

algorithm. First, gray levels are reduced by thresholding to O's for

black or 1's for white. (This is quickly done if one has the

appropriate hardware.) By comparing neighbors and locating the next

edge pixel, the algorithm follows the border counterclockwise for an

outer border and clockwise for an inner one. Since borders may

overlap, provisions are made for backtracking.

In this study, the generality of Rosenfeld's algorithm is un-

necessary, and the algorithm is modified for faster execution. The

modified algorithm (see appendix) is designed to follow outer edges

(no inner edges exist) in the clockwise direction, and reducing the

image is eliminated by combining thresholding with border following.

The only information saved during border following is the y

coordinate of the spot's left-most pixel, L; the y coordinate of the

right-most pixel, R; the x coordinate of the highest pixel, T; and

the x coordinate of the lowest pixel, B. Since the black region

extends a little beyond the approximate border, 2 pixels are sub-

tracted from L and T, while 2 pixels are added to the R and B (see

Figure 8). Thus, the local region of interest (black and inter-

mediate gray) is "framed" for further processing.

Raster scanning resumes at the bottom of the frame and continues

until another black region is found and it is framed. Scanning and

framing continue until the entire image is covered. Thus, the image

is completely segmented and the information needed for feature ex-

traction is available.



















rY

149 151 155 14 1 15 1 1

149 151 155 146 1A3 1SB ISB 1S1


157
150
144
150
146
145
146
150
153
Edge 15&
147
147
145
151


Edge
T

150 ~ A


1441141 149 157 148 156 145 158 145 154
1571150 141 144 158 149 166 150 151 142
156 158 146 160 148 140 148 149 1511144
155 147 150 131 150 150 163 150 158|153
151 160 146 148 144 152 153 148 160 160


1 1


149
147
136


Frame


142 145 146 156 145 160 148 165
153 142 148 157 149 153 148 143
153 155 132 148 144 151 169 164
127 150 154 137 149 152 151 153
60 90 142 154 147 163 146 150'
51 70 52 139 152 150 150 146
46 53 59 91 151 153 153 151
51 45 56 44 135 154 146 154
58 53 50 49 105 141 144 150
50 54 46 47 101 150 143 143
49 55 51 42 103 146 158 1 2
50 50 47 33 137 156 153 13 Edge
48 44 43 82 165 147 153 160
47 50 63 132 151 158 147 148
39 70 141 148 152 148 159 152
143 154 150 145 154 163 159 151
139 145 145 147 154 161 148 149
150 145 156 150 150 151 147 149
149 150 151 153 152 149 157 160
146 163 136 151 154 153 152 153
157 144 144 141 162 146 148 154


Figure 8. Frame of spot in Figure 4.








II.2.4 Image Features


Images contain various types of information known as features,


and, when properly selected, they adequately represent the images.


Here, it is assumed that natural features are intrinsic to the real


scene and artifacts are added as desired to aid subsequent proces-


sing. Features, both natural and artifact, are generally grouped


into three classes: physical, structural, and mathematical [65].


145 156 142 146 150 127
155 156 145 146 150 159
165 156 138 155 146 148
151 147 156 148 138 83
152 146 134 138 66 45
154 140 141 84 56 56
159 143 138 53 61 51
150 150 100 51 48 36
139 15i 96 50 50 44
155 155 108 40 43 46
148 166 137 47 51 47
148 151 153 85 45 50
135 155 159 138 76 55
148 151 162 159 144 82
153 156 153 148 132 148


. . j .










Physical and structural features are, in this case, the arti-

facts shown in Figure 9 plus the pattern they form in the image. For

simplicity, the (single) icon is considered a "special" spot, and all

other spots are assumed uniform in size.






















Spot Icon











Figure 9. Image artifacts. The icon is a "special" spot.






Mathematical features are frame areas, spot centroids, and those

concerning icon identification and rotation. These features allow

extensive integer programming which speeds the majority of the cal-

culations and improves accuracy.











11.2.5 Feature Extraction

Here, feature extraction consists of isolating the artifacts and

computing their desired mathematical features. Since the artifacts

are extracted (by framing) during segmentation, all that remains is

the computation.

Frame area. Frame (integer) area is easily computed using the

information obtained during the segmentation process. The computa-

tion is



A = (R-L)*(B-T) (21)



Since the icon's frame area is much larger than the other spots',

classifying each artifact as "spot" or "icon" is a simple, binary

task.

Spot centroids. The centroids of all spots are used as defini-

tions of spot position. If the gray levels in a spot's frame are

assumed to be "weights," the x and y coordinates of the centroid are

calculated [18] by



B R
E E xI(x,y)
x=T y=L
B R
E E I(x,y)
x=T y=L



(22)
R B
E Z yI(x,y)
y=L x=T
R B
E E I(x,y)
y=L x=T











Henceforth, these summation limits are implied unless otherwise

noted.

How do gray levels, noise, and spot size affect spatial resolu-

tion of the centroid? Here, it is evident (see Figure 8) that merely

summing with gray levels I(x,y) does not yield the proper result

because the white pixels shift the spot's computed centroid away from

the correct value. A transformation of all gray levels in the frame

by the equation



I'(x,y) = I -I(x,y), (23)
w



where I, is the local average white level and I'(x,y) is the noisy

gray level difference (between black and white), gives the approxi-

mate centroid coordinates. And since pixels on the frame edge are

white, their average gives a good, local estimate for I,. The sub-

stitution of I'(x,y) into equations (22) leads to




E E x I'(x,y)
S=xy
E E I'(x,y)
xy


(24)


Z E y I'(x,y)
yx
E E I'(x,y)
yx











To account for noise, assume each gray level in the image con-

sists of the actual (i.e., uncorrupted by noise) value, r(x,y), plus

zero-mean Gaussian noise, n(x,y). Thus,



I'(x,y) = I r(x,y) + n(x,y)
W


I'(x,y) = H(x,y) + n(x,y)


(25)


where H(x,y) = I r(x,y) is the actual gray level difference.

Substituting equation (25) into equations (24) leads to


EExH + ZZxn
xy xy
ZEH + ZEn
xy xy


(26)


EZyH + EZyn
- yx yx
E H + EE
yx yx


If the expected value [63] for the noise is assumed


EjnI(x,y)}


- EEn(x,y)
xy


- ECn(xy)
yx


En(x,y)
x


- En(x,y)
y


(27)










and substituted into equations (26), then the expected centroid is




ZZxH ZxEn
E{X -= xy x y
EEH + EZn
xy xy


ZExH + Ejn}Zx


xy
xy (28)


EZyH + E{n}Ey
E{Y1} yx y
EZH + Enr}
yx



Since E{n}=O, equations (28) become




EExH
E{} =xy
ECH
xy (29)


ElyH
EY} yx
EZH
yx



In general, the summations in equations (27) are not equal to

zero, which introduces random error. But if Zn = 0, then X = E{X},
y
and if Zn = 0, then Y = E{Y}. Thus, the summation process used to
x
compute the centroids greatly reduces, although in general it does

not eliminate, the effect of noise.










To understand better how noise, spot diameter, and number of

gray levels interact to affect spatial resolution, consider the one-

dimension model in Figure 10. This model is essentially a column in

the two-dimensional image (see Figure 8) with frame edges at x=T and

x=B. For simplicity, edge pixels are always located at x=T+2 and

x=B-2. (The extension to the general two-dimension case is straight-

forward.) Similar to equation (26), the centroid calculation is


ExH(x) + Zxn(x)
Sx x
SH(x) + En(x) (30)
x x


_T_


T T+2 T+3 \
line length


I -v


x


B-3 B-2


- F. Frame





Figure 10. One-dimension model.


Assume


a Ii









KT H x=T+2

KHc x=B-2
H(x) =
H(x) H T+3 C
0 otherwise



1 c- m
O
0<(K< <1
o- B-l


and T<6

where Hc, a constant over the line length, is the gray level differ-

ence and Hm is the maximum possible gray level difference (contrast)

between black and white.

The influence of noise and the edge pixels can be described in

terms of resultants. The noise resultant, 6n is the product of

random variables 6 and no, i.e., Exn(x)=6n and En(x)=nc. Dimen-
x x
sionless parameter KT is the fraction of the x=T+2 edge pixel covered

by the line; and KTH is the resultant. Assume the same for
and KBH at edge x=B-2. During digitization, KT and
tized (see Figure 11) to K' and <', respectively. For a given line

segment at a specified location, KT and KB are constants. The sub-

stitution of these values into equation (30) yields


B-3
Z xH +H [T (T+2)+K (B-2)]+6n
Sx=T+3
B-3
Z H(x)+H ( + If x=T+3

B-3
E x = (T+3)+(T+4)+(T+5)+ ... +(B-5)+(B-4)+(B-3)
x=T+3











is added to


3-3
Z x = (B-3)+(B-4)+(B-5)+ ... +(T+5)+(T+4)+(T+3)
x=T+3


then


B-3
2 Z x = (T+B)+(T+B)+(T+B)+ ... +(T+B)+(T+B)+(T+B)
x=T+3


B-3
E x = -(T+B)
x=T+3


where n is the number of terms in the sum.



n = (B-3) (T+3) + 1


Since B-3>T+3 and n>1,


n = B-T-5.


Therefore,


B-3
B3 (B-T-5)(T+B)
E x =
x2T
x=T+3


B-3
E H(x) = (B-T-5)H
x=T+3










Subsequently, equation (30) becomes


(B-T-5)(T+B)Hc+2{6nc+Hc[ S2[(B-T-5)Hc +c+H ( T + )]
c c T B


After dividing the top and bottom by Hc, equation (31) becomes


(B-T-5)(T+B)+2[6c+KT(T+2)+ 2[(B-T-5)+E+KT +

n
where E = H Note, the E ratio is
C
Pratt's [66] signal-to-noise ratio,


similar to the inverse of


H 2
SNR = (c)
o


(33)


1 I


1
I C


Fraction of Edge Pixel covered


Figure 11. The discretization of T and K An increase in He im-
proves spatial resolution of the edges.


(31)


(32)










After studying Figure 10 and equation (32), it is observed

that 6=T or 6=B gives the most noise error while


(B-T-5)(T+B)+2[CE(T+2)+K (B-2)]
6 = T[BB_5+TB (314)
2[(B-T-5)+K +BK


gives none.

If a line length of d=B-T-5 is assumed, equation (32) becomes


(d)(d+2T+5)+2[6+K (T+2)+BKg(d+T+3)]
X = (35)
2[d+S+KT+

If &=0,


(d)(d+2T+5)+2[KT(T+2)+K (d+T+3)]
xJ= (36)
2[d+K +K B



which is the "exact" centroid. If T=0 is assumed and K' and K' are
T B
substituted into equation (35), the resolution error is


(d)(d+5)+2[2KT+K (d+3)] (d)(d+5)+2[6+2K+K'(d+3)]
R = (37)
e 2[d+KT+KB] 2[d+S+<+

The general relationship between E, d, and Re is clearer if

equation (37) is simplified. To assume the line position is a random

variable implies the following expected values:



EIKT} = E{

After some algebra, equation (37) becomes












Re -(d+5-26)
e 2(d++1)


(38)


For E much smaller than d, equation (38) suggests an almost linear

relation between & and R A spatial dependency (perhaps difficult
to verify experimentally) is also indicated.
to verify experimentally) is also indicated.


Icon
Centroid


Figure 12. Icon rotation measured from the vertical.






Icon rotation. Next, the icon's angle of rotation is computed

(see Figure 12) using the second moment of inertia. The analysis is

as follows:











I = EE I'(x,y) d'
xx y
xy


I = EE I'(x,y) d2 (39)
yy x
xy


P = LE I'(x,y) d d
xy xy
xy


where Ixx = moment of inertia about icon's x-axis

I = moment of inertia about icon's y-axis

Pxy = product of inertia about icon's centroid

dx = x distance from pixel to icon centroid

d = y distance from pixel to icon centroid.

Icon rotation is


2P
S= ~Tan -1 x ) (40)
2 I -I
yy xx


The above equation gives the angle to within a constant. For

example, the analysis gives the same angle for both icon orientations

in Figure 13. The difference is resolved by constructing line "A-A"

through the icon centroid and computing the centroids of the upper

and lower segments. Line "A-A" has the form



A = mxA+b


where xA and yA are points on "A-A" with



m = -TAN(e)

and


b = con-mxcon
Icon Icon













SY
x


+6


K


"A" ,'
/


Icon
"A"/


+6


Upper
Segment


Upper
Segment


Icon



Lower
-Segment


Figure 13.


Icon rotation is determined to within a constant by equa-
tion (40).


The centroid for the upper segment is



XA YA
E E xI'(x,y)
S= x=T y=L
XA YA
E E I'(x,y)
x=T y=L


XA YA
E E yI'(x,y)
x=T y=L
XA YA
E Z I'(x,y)
x=T y=L


(41)










and for the lower segment,


B R
E E xI'(x,y)
X=XA Y=YA
XL B R
E E I'(x,y)
X=XA Y=YA
(42)


B R
E E yI'(x,y)
XX A Y=YA
L B R
E E I'(x,y)
X=XA Y-YA


The distances from the icon centroid to the segment centroids are

2 -- 2'
d = U /(X -X ) +(Y -Y con
U U Icon U Icon
(43)


( m on2 2
d Icon ) I+(Y co
L --Icon L Icon


If dU>dL, 9 computed by equation (40) is correct; otherwise 0 must be

shifted by


r9-ir 8>0
e -- (44)
fe+ r 8<0


Thus, a maximum rotation of one revolution is measurable.











Information saved during feature extraction is, of course, frame

area, the centroid coordinates of each spot, the number of spots,

which spot is the icon, and icon rotation. The areas and the cen-

troids are stored as linear arrays in the order the spots were

found. Lastly, the problem of sorting this raw information and using

it to match spots in the before- and after-motion images is solved

using pattern recognition principles.

11.2.6 Known Patterns

In this study, patterns are classified as either "known" or

"unknown." Known patterns contain structural features (natural,

artifact, or both) that are recognizable before and after motion.

Unknown patterns are patterns which never have recognizable structure

or those in which features are obscured beyond recognition. Known

patterns containing only artifacts are considered here.

Artifacts and their arrangement are selected to improve segmen-

tation, feature extraction, mapping, and subsequent numerical

analysis. Examples of simple, easily analyzed patterns are shown in

Figure 14. These patterns are fine for small rotations and are

easily recognized because the relative order of the spots does not

change, e.g., the upper left spot remains in the same position--

relative to the other spots--during motion. If rotation is 45

degrees or more for the square patterns and 90 degrees or more for

the others, the relative order is lost. The patterns in Figure 15

allow spot identification after rotations up to one revolution. Once

the patterns in the undeformed and deformed images have been

















66 56555

ewes .me

* sue...

* @505mg

* wsswg

S 5 66 m

ewe....









6 6



) 6 5


Figure 14. Patterns that are easily recognized and analyzed. Frames

are for reference and not part of the pattern.


*ee ee



* U U U






0C0 eeO









S

sew.

wee

sees


Figure 15. Patterns of Figure 14 with icons. The icons are located

for convenience.


000000 0


* U U U


00000 *0


* 3 6 S


@6 *5e66

ewe....

66 @6 Se

* *..L@ e U

sew....

.me....

ewe....


* C 9

* w

* S S

* 6 5











recognized, subsequent analysis is simplified (especially for the

Lagrangian description) because the undeformed spacing is approxi-

mately constant.

11.2.7 Syntactic Pattern Recognition

Pattern recognition is usually divided into two classes: deci-

sion-theoretic and syntactic. In the first class, a pattern is

represented by feature vectors and recognized by a stochastic parti-

tioning of feature space [67-69]. Syntactic recognition uses pattern

structure and linguistic rules for analysis [70,71].

Syntactic recognition assumes a pattern is described by a

language which contains valid sentences composed from a finite alpha-

bet or set of symbols. The language is governed by "pattern grammar"

or set of grammar rules. In context-free languages, pattern grammar

is unrestricted. Context-sensitive languages restrict pattern

grammar on the number of sentences, sentence construction, or both.

Here, the pattern language is context-sensitive and has only one

sentence which is composed of a fixed number of ordered symbols. The

alphabet symbols are the spots' relative positions whose order is

determined in the undeformed image. A spot's context is determined

by its 0,2,4,6 (or alternatively by 0 through 7) neighbors. For

example, (2,2) has neighbors (1,2), (2,1), (3,2), and (2,3). An

attempt to map (2,2) in the undeformed image to, say (1,1), in the

deformed is "out of context" with the neighbors (see Figure 16).

Thus, correct pattern mapping is assured if every spot maps in con-

text with its neighbors.





















*
(1,1) (1,2) (1.3) (1,4) (1,5)
(2. 1) (2,2) (2.3) (2.4) (2.5)
S ** (3.1) (3.2) (3.3) (3.4) (3.5)
0 0 4 (4,1) (4,2) (4.3) (4,4) (4.5)
(5.) (5 ) (52) (5.3) (5,4) (5.5)





Pattern One-sentence Language






Figure 16. A pattern and its context-sensitive language. Note the
alphabet (symbols) and context in the language.





Higher dimensional pattern grammars [70-72] provide the capabil-

ity to formally describe, recognize, and map patterns similar to

those in Figure 15. Since a record of each spot and its neighbors

must be maintained and manipulated, generating and parsing formal

grammar for large patterns becomes an intense process. If possible,

trading some of the formal elegance for an economic approximation is

highly desirable.

11.2.8 Approximation of Syntactic Pattern Mapping

Fortunately for many motions, spot relative positions change

very little, which allows context to be determined from spot






43



"addresses." If both images are carefully divided into a number of

addresses subregionss), then a spot has the same address before and

after motion--or nearly so. The determination of a spot's address is

a simple calculation involving centroid coordinates, and recognition

is reduced to merely arranging the spot addresses in proper order.

Thus, context is recognized.

"Offsets" are the spot addresses relative to the icon address.

All offsets are defined according to an icon address of 0 (see Figure

17).


Image
Border


- --K--------


Spot
Addresses
*
-18 -10 -2 6 14

*
-17 -9 -1 7 15

01 *
-16 -8 0 8 16

*
-15 -7 1 9 17

*0
1 -14 -6 2 10 18

"Offee se


Figure 17. Image addresses and offsets.
image edges.


Some addresses overlap









Pattern geometry affects the upper limits of measurable strain

because large strain displacements may change the order of the off-

sets in the deformed image. Spreading spots further apart (in the

undeformed image) increases the size of a spot's subregion. A larger

subregion allows the spot more strain displacement before any change

in offset order occurs.

The details are best illustrated by example (see Figure 18). It

is assumed that the undeformed image is analyzed first and, with one

exception, both images receive the same preliminary analysis.


* *
* emiS* *






Undeformed


U
U
0


+ o
U+

0
U
eo




Deformed


Figure 18. A simple example of motion (au./3X.=0.05, u=v=-10 pixels,
e=1200, E..=0.0525). The origin is in the upper left
corner of 1ach image.










Successful mapping is initially assumed and later verified by four

final checks.

First, the image is segmented and the features are extracted.

Scanning is by column and the spot centroids are stored in the order


they were found. (See columns marked "Raw" in Tables 1 and 2.)


The


icon is identified, and it's rotation is also computed.




Table 1. Unordered spot coordinates in the undeformed image.


Reference


From Icon


X Y X Y X Y


58.0101
93.0079
163.0518
197.9896
127.9832
58.0481
93.0216
197.9616
127.9396
163.0107
128.7025
57.9599
93.0284
198.0660
63.0010
92.9964
127.9581
162.9653
197.9768
57.9856
58.0635
92.9969
128.0099
163.0027
197.9936


57.9660
58.0056
57.9950
58.0041
57.9801
93.0576
92.9791
93.0132
92.9724
92.9915
127.9873
128.0263
128.0002
128.0575
127.9894
163.0297
162.9798
163.0147
162.9531
162.9875
198.0315
198.0169
198.0273
198.0136
197.9982


57.4541
92.4517
162.4955
197.4332
127.4271
57.4187
92.3923
197.3321
127.3103
162.3813
128.0000
57.2575
92.3260
197.3632
162.2984
92.2207
127.1824
162.1896
197.2011
57.2101
57.2148
92.1481
127.1610
162.1538
197.1446


57.8310
57.9438
58.0797
58.1617
57.9914
92.9227
92.9173
93.1757
92.9836
93.0760
128.0000
127.8911
127.9383
128.2151
128.0738
162.9676
162.9909
163.0989
163.1105
162.8523
197.8964
197.9548
198.0384
198.0978
198.1556


-70.5459
-35.5483
34.4955
69.4332
-0.5729
-70.5813
-35.6077
69.3321
-0.6897
34.3813
0.0000
-70.7425
-35.6740
69.3632
34.2984
-35.7793
-0.8176
34.1896
69.2011
-70.7899
-70.7852
-35.8519
-0.8390
34.1538
69.1446


-70.1690
-70.0562
-69.9203
-69.8383
-70.0086
-35.0773
-35.0827
-34.8243
-35.0164
-34.9240
0.0000
-0.1089
-0.0617
0.2151
0.0738
34.9676
34.9909
35.0989
34.1105
34.8523
69.8964
69.9548
70.0384
70.0978
70.1556


NOTE: Units are pixels.


"Raw" coordinates are the spot centroids


in the image. "Reference" and "From Icon" values are computed by
equations (46) and (47), respectively. See Figure 18.










Table 2. Unordered spot coordinates in the deformed image.


Reference


From Icon


X Y X Y X Y


98.5336
131.2724
78.6761
163.9343
111.3736
196.6288
58.7113
144.0612
229.4016
91.4833
176.7262
123.7393
38.9059
209.4617
71.5500
156.8817
18.9873
104.3150
189.5733
51.6854
136.9841
84.3163
169.6761
117.1079
149.7807


33.4158
50.2682
64.3727
67.1092
81.1960
84.0388
95.2677
98.0816
100.8559
112.1065
114.9435
129.6776
126.2522
131.8162
143.1225
145.8710
157.2419
160.0107
162.8025
174.0585
176.8253
190.9335
193.7303
207.7552
224.6135


51.9634
53.0816
88.4532
54.2228
89.5630
55.4298
124.9335
90.7333
56.5016
126.0252
91.8919
128.0000
161.4258
93.0300
162.6010
128.3497
197.9716
163.7403
129.5597
199.0750
164.8545
200.2602
166.0404
201 .3280
202.4800


192.1880
155.3833
196.7857
118.6530
160.0309
81 .8551
201.5070
123.2581
45.0350
164.6782
86.5161
128.0000
206.0458
49.7056
169.3190
91.1147
210.6846
132.4753
54.3186
173.9321
95.7498
137.2151
58.9647
100.3761
63.6284


-76.0366
-74.9184
-39.5468
-73.7772
-38.4370
-72.5702
-3.0665
-37.2667
-71.4984
-1.9748
-36.1081
0.0000
33.4258
-34.9700
34.6010
0.3497
69.9716
35.7403
1.5597
71 .0750
36.8545
72.2602
38.0404
73.3280
74.4800


64.1880
27.3833
68.7857
-9.3470
32.0309
-46.1449
73.5070
-4.7419
-82.9650
36.6782
-41.4839
0.0000
78.0458
-78.2944
41.3190
-36.8853
82.6846
4.4753
-73.6814
45.9321
-32.2502
9.2151
-69.0353
-27.6239
-64.3716


NOTE: Units are pixels.


"Raw" coordinates are the spot centroids


in the image. "Reference" and "From Icon" values are computed by
equations (46) and (47), respectively. See Figure 18.



All spots are moved into a "reference position" by translation

and rotation (see Figure 19). Translation is accomplished by adding

X to the X and YT to the Y of each spot; X and YT are



X = X -X
T cp Icon

(45)


Y = Y -Y
T cp Icon


I











where X and Y are the coordinates of the icon in reference pos
cp cp
tion (typically the image center). Rotation for each spot

computed from the x and y distance between it and the icon. T

computation is


X = X-X
p Icon



Y = Y-Y
p Icon



p = X + Y
P P


-1 p
_ = Tan (-)
z X
P
p


XRef = X+p[cos( z-0)-cos z ]+XT



YRe = Y+p[sin(w -6)-sin w ]+Y
Ref z z T


where XRef and YRef are the new coordinates of each

reference position. The results of translation and

shown in Tables 1 and 2 under the "Reference" column.

Position relative to the icon, i.e., distance from


spot in

rotation



the icon,


XRel XRef XIcon



Y =Y -YIcon
Rel Ref Icon


(46)


the

are



is


(47)






48












0


-


) e *
S*



Deformed Image Reference Position






Figure 19. The deformed image and its reference position.





The distances from the icon are shown in Tables 1 and 2 under the

columns marked "From Icon." Although all spots have been located

relative to the icon, recognizing the pattern they form and arranging

them in proper order (by address) is not always easy--especially if

rotation is more than a few degrees.

Address computation is based on "learning" the undeformed

image. The spots nearest and farthest from the icon row and column

are located in the undeformed image and used to calculate the address

"offset," Aoffset, by the following:










X = MAX(X)
max



m = MIN(X)



Y = MAX(Y)
max


Y
min


= MIN(Y)


X
N = max
N = NINT( )
x X
mln

Y
max
N = NINT( )
Smin


X -N (X )
SX + max x min
X = X +
space min N +1
X

Y -N (Y )
Ymax y min
Y = Y +
space min N +1


N
N = NINT( ) + 1
shift X
space


Xel YRel
Offset = NINT( ) + NINT( )(N )
offset X space sift
space space


where Xmax


Xmin

Ymax

Ymin

N =


= maximum

= minimum

= maximum

= minimum

number of


X centroid

X centroid

Y centroid

Y centroid

rows above or below icon


(48)


(49)



(50)










Ny = number of columns left or right of icon

Nrows = number of rows in the image

Xspac = approximate row spacing

Yspace = approximate column spacing

Nshift = number of possible columns in image.

Function NINT rounds numbers to the nearest integer value. Values

Space' space' and Nshift are saved from the undeformed image and

used in the deformed.

Fortunately, arranging the spots in proper sequence is a simple

matter of bubble sorting [73] the offsets in ascending order. By

equation (50), offsets (see column "Offset" in Tables 3 and 4)

increase down columns and left to right. Column "Key" is the order

in which the spots were found. Note, the two-dimension pattern

context is now represented by a one-dimension array, XRel, which

corresponds to the ordered spot addresses.

Recognition of the two-dimension context is accomplished by

determining the row-column order of the spots. The left-most column

is assumed to begin with the lowest offset and continue with ascend-

ing offset until the X coordinate decreases. Column 2 begins with

the next highest offset and continues until the X coordinate again

decreases. The process continues until the offset array is exhaust-

ed. By incrementing row and column counters, the relative spot order

is determined which establishes context (see Tables 5-8).

Until now, both undeformed and deformed images were processed

the same (except for learning the undeformed image) and assumed to

have the same context. This assumption is verified by the following












Table 3. Ordered spot coordinates in the undeformed image.


Reference


From Icon


Y

57.8310
57.9438
57.9914
58.0797
58.1617
92.9227
92.9173
92.9836
93.0760
93.1757
127.8911
127.9383
128.0000
128.0738
128.2151
162.8523
162.9676
162.9909
163.0989
163.1105
197.8964
197.9548
198.0384
198.0978
198.1556


X Y


Offset

-18
-17
-16
-15
-14
-10
-9
-8
-7
-6
-2
-1
0
1
2
6
7
8
9
10
14
15
16
17
18


Key

1
2
5
3
4
6
7
9
10
8
12
13
11
15
14
20
16
17
18
19
21
22
23
24
25


NOTE: Units are pixels.


Offsets are computed by equation (50).


-70.5459
-35.5483
-0.5729
34.4955
69.4332
-70.5813
-35.6077
-0.6897
34.3813
69.3321
-70.7425
-35.6740
0.0000
34.2984
69.3632
-70.7899
-35.7793
-0.8176
34.1896
69.2011
-70.7852
-35.8519
-0.8390
34.1538
69.1446


-70.1690
-70.0562
-70.0086
-69.9203
-69.8383
-35.0773
-35.0827
-35.0164
-34.9240
-34.8243
-0.1089
-0.0617
0.0000
0.0738
0.2151
34.8523
34.9676
34.9909
35.0989
35.1105
69.8964
69.9548
70.0384
70.0978
70.1556


X

57.4541
92.4517
127.4271
162.4955
197.4332
57.4187
92.3923
127.3103
162.3813
197.3321
57.2575
92.3260
128.0000
162.2984
197.3632
57.2101
92.2207
127.1824
162.1896
197.2011
57.2148
92.1481
127.1610
162.1538
197.1446


Keys are the order in which the spots were found. Undeformed
spacing is 35 pixels in x and y. See Figure 18.










Table 4. Ordered spot coordinates in the deformed image.



Reference From Icon

Offset Key X Y X Y

-18 9 56.5016 45.0350 -71.4984 -82.9650
-17 14 93.0300 49.7056 -34.9700 -78.2944
-16 19 129.5597 54.3186 1.5597 -73.6814
-15 23 166.0404 58.9647 38.0404 -69.0353
-14 25 202.4800 63.6284 74.4800 -64.3716
-10 6 55.4298 81.8551 -72.5702 -46.1449
-9 11 91.8919 86.5161 -36.1081 -41.4839
-8 16 128.3497 91.1147 0.3497 -36.8853
-7 21 164.8545 95.7498 36.8545 -32.2502
-6 24 201.3280 100.3761 73.3280 -27.6239
-2 4 54.2228 118.6530 -73.7772 -9.3470
-1 8 90.7333 123.2581 -37.2667 -4.7419
0 12 128.0000 128.0000 0.0000 0.0000
1 18 163.7403 132.4753 35.7403 4.4753
2 22 200.2602 137.2151 72.2602 9.2151
6 2 53.0816 155.3833 -74.9184 27.3833
7 5 89.5630 160.0309 -38.4370 32.0309
8 10 126.0252 164.6782 -1.9748 36.6782
9 15 162.6010 169.3190 34.6010 41.3190
10 20 199.0750 173.9321 71.0750 45.9321
14 1 51.9634 192.1880 -76.0366 64.1880
15 3 88.4532 196.7857 -39.5468 68.7857
16 7 124.9335 201.5070 -3.0665 73.5070
17 13 161.4258 206.0458 33.4258 78.0458
18 17 197.9716 210.6846 69.9716 82.6846

NOTE: Units are pixels. Offsets are computed by equation (50).


Keys are the order in which the spots were found.


See Figure 18.














Table 5. Spot distance from origin in undeformed image.



COL 1 2 3 4 5

ROW 1 X: 58.0101 58.0481 57.9599 57.9856 58.0635
Y: 57.9660 93.0576 128.0263 162.9875 198.0315

COL 1 2 3 4 5

ROW 2 X: 93.0079 93.0216 93.0284 92.9964 92.9969
Y: 58.0056 92.9791 128.0002 163.0297 198.0169

COL 1 2 3 4 5

ROW 3 X: 127.9832 127.9396 128.7025 127.9581 128.0099
Y: 57.9801 92.9724 127.9873 162.9798 198.0273

COL 1 2 3 4 5

ROW 4 X: 163.0518 163.0107 163.0010 162.9653 163.0027
Y: 57.9950 92.9915 127.9894 163.0147 198.0136

COL 1 2 3 4 5

ROW 5 X: 197.9896 197.9616 198.0660 197.9763 197.9936
Y: 58.0041 93.0182 128.0575 162.9531 197.9982

NOTE: Units are pixels. The origin is the upper left corner of the
image. See Figure 18.














Table 5. Spot distance from origin in deformed image.


COL 1 2 3 4 5

ROW 1 X: 229.4016 196.6288 16-3.9343 131.2724 98.5336
Y: 100.8559 84.0388 67.1092 50.2682 33.4158

COL 1 2 3 4 5

ROW 2 X: 209.4617 176.7262 144.0612 111.3736 78.6761
Y: 131.8162 114.9435 98.0816 81.1960 64.3727

COL 1 2 3 4 5

ROW 3 X: 189.5733 156.8817 123.7393 91.4833 58.7113
Y: 162.8025 145.8710 129.6776 112.1065 95.2677

COL 1 2 3 4 5

ROW 4 X: 169.6761 136.9841 104.3150 71.5500 38.9059
Y: 193.7303 176.8253 160.0107 143.1225 126.2522

COL 1 2 3 4 5

ROW 5 X: 149.7807 117.1079 84.3163 51.6854 18.9873
Y: 224.6135 207.7552 190.9335 174.0585 157.2419

NOTE: Units are pixels. The origin is the upper left corner of the
image. See Figure 18.













Table 7. Spot distance from icon in undeformed image.


COL 1 2 3 4 5

ROW 1 X: -70.5459 -70.5813 -70.7425 -70.7899 -70.7852
Y: -70.1690 -35.0773 -0.1089 34.8523 69.8964

COL 1 2 3 4 5

ROW 2 X: -35.5483 -35.6077 -35.6740 -35.7793 -35.8519
Y: -70.0562 -35.0827 -0.0617 34.9676 69.9548

COL 1 2 3 4 5

ROW 3 X: -0.5729 -0.6897 0.0000 -0.8176 -0.8390
Y: -70.0086 -35.0164 0.0000 34.9909 70.0384

COL 1 2 3 4 5

ROW 4 X: 34.4955 34.3813 34.2984 34.1896 34.1538
Y: -69.9203 -34.9240 0.0738 35.0989 70.0978

COL 1 2 3 4 5

ROW 5 X: 69.4332 69.3321 69.3632 59.2011 69.1446
Y: -69.8383 -34.8243 0.2151 35.1105 70.1556

NOTE: Units are pixels. Spacing is 35 pixels in x and y. See
Figure 18.













Table 8. Spot distance from icon in deformed image.


COL 1 2 3 4 5

ROW 1 X: -71.4984 -72.5702 -73.7772 -74.9184 -76.0366
Y: -82.9650 -46.1449 -9.3470 27.3833 64.1880

COL 1 2 3 4 5

ROW 2 X: -34.9700 -36.1081 -37.2667 -38.4370 -39.5468
Y: -78.2944 -41.4839 -4.7419 32.0309 68.7857

COL 1 2 3 4 5

ROW 3 X: 1.5597 0.3497 0.0000 -1.9748 -3.0665
Y: -73.6814 -36.8853 0.0000 36.6782 73.5070

COL 1 2 3 4 5

ROW 4 X: 38.0404 36.8545 35.7403 34.6010 33.4258
Y: -69.0353 -32.2502 4.4753 41.3190 78.0458

COL 1 2 3 4 5

ROW 5 X: 74.4800 73.3280 72.2602 71.0750 69.9716
Y: -64.3716 -27.6239 9.2151 45.9321 82.6845

NOTE: Units are pixels. See Figure 18.










checks:

1. same number of rows in each image

2. same number of columns in each image

3. same number of spots in each image

4. rows times columns equal number of spots.

Of course, if any check fails, the images are "out of con-

text." If strains are large enough, context similarities between

images cannot be recognized and an error message is printed. Assum-

ing large, local rotation does not occur in the material, recognizing

similar context between the two patterns that does not exist is very

remote. Thus if no check fails, the approximation algorithm estab-

lishes a one-to-one relationship, or "context mapping," between spots

in the two images.
















CHAPTER III
IMAGE GENERATION AND ANALYSIS



III.1 Synthetic Images

Computer generated images, or "synthetic" images, were used as

the primary development tool and to study the influence of various

parameters. The options data file shown in Table 9 was used to gen-

erate these 256X256 pixel images.

Motion is limited to affine transformations plus rigid-body

rotation about the image center. The transformations are



x = c X + c2Y + 3

(51)

y = c4X + c5Y + 6



There are three image output options available. The "BINARY"

option is the machine representation of the image. "DECIMAL" is the

machine version converted to a readable form (see, for example,

Figure 4b). "PLOT" displays a thresholded image, suitable for hard

copy, on a graphics terminal (see, for example, Figure 15).

"U123.DAT," "NIMG.DAT," "D123.DAT," and "DIMG.DAT" are user-defined

file names. The image generator automatically reads the input file

and generates the specified output thereby allowing the user a choice

of interactive or batch processing.










Table 9. Input file for synthetic image generator.



IMAGE GENEREATOR INPUT DATA


SPOT GEOMETRY


RADIUS XSPACING
4.00 35.00


PATTERN SIZE

XCOLS YROWS
5 5


YSPACING
35.00


GRAY LEVELS

BLACK WHITE
00000 00200


NOISE DATA


DISTRIBUTION TYPE: N


NORMAL


STANDARD DEVIATION: 007.000


UNIFORM


IMAGE MOTION

MOTION TYPE: 2

1--POLYNOMIAL ONLY
2--POLYNOMIAL AND ROTATION
3--ROTATION ONLY


POLYNOMIAL


X Y TRANSLATION


x: +1.05000 +0.05000
y: +0.05000 +1.05000

ROTATION (DEGREES): +120.000


-010.0000
-010.0000


IMAGE OUTPUT


FORMAT

BINARY:
DECIMAL:
PLOT:


UNDEFORMED

U123.DAT
NIMG.DAT
*SCREEN*


DEFORMED

D123.DAT
DIMG.DAT
*SCREEN*


Y/N

Y
N
N











III.2 Image Analysis

Similar to image generation, image analysis begins with an op-

tions file which is shown in Table 10. If "USE THESE FILE NAMES?" is

answered no, the user is automatically prompted for the file name.

Since no other information is required by the program, the user is

free to choose interactive or batch processing.






Table 10. Image analysis options file.



******** IMAGE ANALYSIS INPUT/OUTPUT OPTIONS ********


INPUT SPECIFICATIONS

IMAGE FILE NAME:
USE THESE FILE NAMES?:


IMAGE OUTPUT

UNORDERED SPOT DATA:
DISTANCE FROM ORIGIN:
DISTANCE FROM ICON:


DISPLACEMENT ANALYSIS

TOTAL DISPLACEMENT:
DISPLACEMENT GRADIENTS:
LAGRANGIAN STRAIN:
TAYLOR SERIES STRAIN:


MOTION ANALYSIS

DEFORMATION GRADIENT:
GREEN DEFORM. TENSOR:
RIGHT STRETCH TENSOR:
ROTATION TENSOR:
LAGRANGIAN STRAIN:


UNDEFORMED
-----+------
U123.DAT
N


UNDEFORMED
-----+------
Y
Y
Y


DEFORMED
-----+------
D123.DAT
N


DEFORMED
------------
Y
Y
Y


OUTPUT

Y
Y
Y
Y


OUTPUT

Y
Y
Y
Y
Y









After segmentation, feature extraction, and context recognition,

automatic processing continues with motion analysis. The motion of

each spot is resolved and the displacement is computed by equation

(3) (see Table 11). Next, the deformation and displacement gradients

at each spot are computed by differentiating the functions used to

describe motion and displacement.



Table 11. Displacement of each spot.



COL 1 2 3 4 5

ROW 1 X: 171.3916 138.5808 105.9744 73.2868 40.4701
Y: 42.8899 -9.0188 -60.9171 -112.7193 -164.6158

COL 1 2 3 4 5

ROW 2 X: 116.4538 83.7046 51.0328 18.3772 -14.3208
Y: 73.8106 21.9643 -29.9186 -81.8337 -133.6442

COL 1 2 3 4 5

ROW 3 X: 61.5900 28.9421 -4.9632 -35.4747 -69.2986
Y: 104.8223 52.8985 1.6903 -50.8733 -102.7596

COL 1 2 3 4 5

ROW 4 X: 6.6243 -26.0267 -58.6861 -91.4154 -124.0968
Y: 135.7353 83.8338 32.0212 -19.8922 -71 .7614

COL 1 2 3 4 5

ROW 5 X: -48.2088 -80.8538 -113.7497 -146.2914 -179.0063
Y: 166.6094 114.7370 62.8761 11.1054 -40.7563

NOTE: Units are pixels. Compare images in Figure 18. Displacements
are computed using the data in Tables 5 and 6 in equation (3).


The choice of a mapping model depends, of course, on the

application. Intuition, a priori knowledge, elasticity theory,










numerical efficiency, etc. are used to select the approximating func-

tions. In this study, simple polynomials, linear or second order,

adequately approximate the motion and displacement in the synthetic

and real images. The models are linear



x = 11X + c21 + 31Y



y = c12X + c22 32Y

(52)


u = c13X + c23 c33Y


v 0 C14X + c24 + C34Y



and second order


2 2
x= 11 + X c+ 31 + c41 + cY + + c XY
11 21 31 41 51 01


2 2
y = C12 + c22X + 32 + cY2 + 52Y + c6XY



u = 13X + c23X + c33 + cY cY Y + c XY
13 23 33 143 53 63


(53)


2 2
v = c1X2 c2X + c3 + 44Y + c54Y + c64XY



These models are fitted to a small neighborhood of spots and

differentiated to obtain the displacement and deformation






63


gradients. Typically, the neighborhoods are 3X3, 5X5, or 7X7 and the

gradients are computed for the center spot. Exceptions are the spots

near the edge of the pattern and their gradients are computed like a

forward-backward difference.

For both models, the coefficients are determined by least

squares evaluation. Based on a modified form of the Conte and Deboor

[74] notation, the normal equations are



T c = f (54)



where




X X2 ." XN

= 1 ... Linear

Y1 Y2 N




(55)

2 2 2
X X2 ... XN
1 2 N
X1 X2 XN

1 1 1
2 2 2 Second
Y Y ... Y Order

Y Yx ... YN

X1Y1 X2Y2 ... XNYN











Linear


c 01 c14

c 31 c 34_
L_ _3 j


(56)


C = i.


61


I:
x 1

f =

xN
"N


Second
order


1 u061

Y1 Ul V1


and N is the number of spots in the neighborhood.



T
S= A


and


Equations (54) become


Ac = Z


which are solvable by Gauss-Jordan elimination.


(57)


(58)


f = z










Hornbeck [75] points out Gauss-Jordan elimination may be used to

solve equations (58), but unfortunately the set is extremely

ill-conditioned. Taking his advice, double precision arithmetic is

used to cope with the large variation in the magnitudes of the co-

efficients in any given row. In addition to his suggestions, 0 and f

are scaled before solution and the c's are reconverted during

gradient computation.

For a variety of programming reasons, solution of equations (58)

is not performed directly by Gauss-Jordan elimination. Instead, the

inverse of A is computed using Hornbeck's [75] Gauss-Jordan matrix

inversion subroutine, with the addition of appropriate error traps,

and c is solved by



-1
c = A Z (59)



How accurate are the values of c and how well do equations (52)

or (53) fit the data? Debugging experience indicates the number of

reliable digits (numerically speaking, not experimentally) runs in

the teens. In this study the solutions are known, but if they were

unknown, goodness-of-fit and the contribution of each term could be

easily tested by the methods suggested by Miller and Freund [63].

The example of Figure 18 is continued with the calculation of

c. Assume the linear model of equation (52) and a 3X3 neighborhood

centered at spot (3,3). A substitution of data from Table 5 into

equation (55) produces













T
I =


93.0216
93.0284
92.9964
127.9396
128.7025
127.9581
163.0107
163.0010
162.9653


1 .0000
1.0000
1.0000
1.0000
1 .0000
1.0000
1 .0000
1.0000
1.0000


92.9791
128.0002
163.0297
92.9724
127.9873
162.9798
92.9915
127.9894
163.0147


(60)


Likewise, data from Tables 6 and 11 in equation (57) lead to


176.7262
144.0612
111 .3736
156.8817
123.7393
91.4833
136.9841
104.3150
L71 .5500


114.9435
98.0816
81.1960
145.8710
129.6776
112.1065
176.8253
160.0107
143.1225


83.7046
51.0328
18.3772
28.9421
-4.9632
-36.4747
-26.0267
-58.6861
-91 .4154


21.9643
-29.9186
-81.8337
52.8985
1 .6903
-50.8733
83.8338
32.0212
-19.8922


(61)


The solution for c is


-0.56849
c = 316.52130
_-0.93425


0.88427
77.50840
-0.48175


-1.56849
316.52130
-0.93425


0.88427
77.50840
-1.48165


(62)


The motion functions (i.e., the "mathematical" maps) in equa-

tions (52) become



x = -0.56849X + 316.52130 0.93425Y


(63)


y = 0.88427X + 77.50840 0.48175Y


I









and the displacement maps in equations (52) become



u = -1.56849X + 316.52130 0.93425Y


(64)


v = 0.88427X + 77.5084 1.48175Y


The deformation and displacement gradients, computed by equa-

tions (2) and (4), respectively, become


-0.56849

0.88427



-1 .56849

0.88427


-0.93425

-0.48175



-0.93425

-1.48175]


(65)


(66)


Strain, computed by equations (11) or (12), is


.05255

0[.05256


0.05256

0.05246]


(67)


By the Taylor series method of equation (13), strain is


0.05124
0.04763


0.0476 3

0.05115


(68)






68



Strain computed according to small displacement theory is, in

this case, quite inaccurate. By equations (14),


-1 .56849
e =
-0.02499



Rigid-body rotation is computed

the Green deformation tensor is




[ .10509
0.10511


-0.02499

-1.48175_


(69)


as follows. By equation (8),


0.10511

1.10491


(70)


The eigenvalues of C are


1 .21009
S= .00000
~p 0.00000


0.00000

0.99990


(71)


1/2
and C 2 is
-p


1/2
~p


1 .10004

0.00000


0.00000

0.99995


(72)


1/2
Since U = C then
-p -p



1.05000

0.05005


0.05005

1.05000


(73)





69



Rotation tensor R, computed by equation (17), is



F-0.50016 -0.86596

0.86596 -0.50011



The remainder of the output for the example in Figure 18 is

shown in Tables 12 through 19.















Table 12. Displacement gradients.


COL 1 2 3 4 5

ROW 1 au/ax: -1.56849 -1.56849 -1.56849 -1.56849 -1.56849
3v/ax: 0.88427 0.88427 0.88427 0.88427 0.88427
au/3Y: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
av/aY: -1 .48175 -1.48175 -1.48175 -1 .48175 -1.48175

COL 1 2 3 4 5

ROW 2 3u/aX: -1.56849 -1.56849 -1.56849 -1.56849 -1.56849
av/ax: 0.88427 0.88427 0.88427 0.88427 0.88427
3u/3Y: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
3v/3Y: -1.48175 -1.48175 -1.48175 -1.48175 -1.48175

COL 1 2 3 4 5

ROW 3 3u/aX: -1.56849 -1.56849 -1.56849 -1.56849 -1.56849
av/3X: 0.88427 0.88427 0.88427 0.88427 0.88427
3u/3Y: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
av/aY: -1.48175 -1.48175 -1.48175 -1.48175 -1.48175

COL 1 2 3 4 5

ROW 4 3u/3X: -1.56849 -1.56849 -1.56849 -1.56849 -1.56849
av/ax: 0.88427 0.88427 0.88427 0.88427 0.88427
au/3Y: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
av/3Y: -1.48175 -1.48175 -1.48175 -1.48175 -1.48175

COL 1 2 3 4 5

ROW 5 3u/3X: -1.56849 -1.56849 -1.56849 -1.56849 -1.56849
av/ax: 0.88427 0.88427 0.88427 0.88427 0.88427
au/3Y: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
3v/3Y: -1.48175 -1.48175 -1.48175 -1.48175 -1.48175

NOTE: Correct values are 3u/3X = -1.56830, 3v/3X = 0.88433,


3u/3Y = -0.93433, 3v/3Y =


-1 .48170.











Table 13. Lagrangian strain from displacement gradients.



COL 1 2 3 4 5

ROW 1 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 2 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 3 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 4 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 5 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

NOTE: Units are micro inches-per-inch. Correct values are E -
52,500.












Table 14. Strain from Taylor series.


COL 1 2 3 4 5

ROW 1 E11: 51241. 51241. 51241. 51241. 51241.
E12: 47634. 47634. 47634. 47634. 47634.
E22: 51151. 51151. 51151. 51151. 51151.

COL 1 2 3 4 5

ROW 2 E11: 51241. 51241. 51241. 51241. 51241.
E12: 47634. 47634. 47634. 47634. 47634.
E22: 51151. 51151. 51151. 51151. 51151.

COL 1 2 3 4 5

ROW 3 E11: 51241. 51241. 51241. 51241. 51241.
E12: 47634. 47634. 47634. 47634. 47634.
E22: 51151. 51151. 51151. 51151. 51151.

COL 1 2 3 4 5

ROW 4 E11: 51241. 51241. 51241. 51241. 51241.
E12: 47634. 47634. 47634. 47634. 47634.
E22: 51151. 51151. 51151. 51151. 51.151.

COL 1 2 3 4 5

ROW 5 E11: 51241. 51241. 51241. 51241. 51241.
E12: 47634. 47634. 47634. 47634. 47634.
E22: 51151. 51151. 51151. 51151. 51151.

NOTE: Units are micro inches-per-inch. Correct values are E1 = E22
= 51,190, E12 = E21 = 47,582.












Table 15. Deformation gradient.


COL 1 2 3 4 5

ROW 1 F11: -0.56849 -0.56849 -0.56849 -0.56849 -0.56849
F12: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
F21: 0.88427 0.88427 0.88427 0.88427 0.88427
F22: -0.48175 -0.48175 -0.48175 -0.48175 -0.48175

COL 1 2 3 4 5

ROW 2 F11: -0.56849 -0.56849 -0.56849 -0.56849 -0.56849
F12: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
F21: 0.88427 0.88427 0.88427 0.88427 0.88427
F22: -0.48175 -0.48175 -0.48175 -0.48175 -0.48175

COL 1 2 3 4 5

ROW 3 F11: -0.56849 -0.56849 -0.56849 -0.56849 -0.56849
F12: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
F21: 0.88427 0.88427 0.88427 0.88427 0.88427
F22: -0.48175 -0.48175 -0.48175 -0.48175 -0.48175

COL 1 2 3 4 5

ROW 4 F11: -0.56849 -0.56849 -0.56849 -0.56849 -0.56849
F12: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
F21: 0.88427 0.88427 0.88427 0.88427 0.88427
F22: -0.48175 -0.48175 -0.48175 -0.48175 -0.48175

COL 1 2 3 4 5

ROW 5 F11: -0.56849 -0.56849 -0.56849 -0.56849 -0.56849
F12: -0.93425 -0.93425 -0.93425 -0.93425 -0.93425
F21: 0.88427 0.88427 0.88427 0.88427 0.88427
F22: -0.48175 -0.48175 -0.48175 -0.48175 -0.48175


NOTE: Correct
0.88433, F22 =


values are
-0.48170.


F11 = -
0.56830,
Fl2 = -
0.93433,
?21 "













Table 16. Green deformation tensor


COL 1 2 3 4 5

ROW 1 C11: 1.10511 1.10511 1.10511 1.10511 1.10511
C12: 0.10511 0.10511 0.10511 0.10511 0.10511
C22: 1.10492 1.10492 1.10492 1.10492 1.10492

COL 1 2 3 4 5

ROW 2 C11: 1.10511 1.10511 1.10511 1.10511 1.10511
C12: 0.10511 0.10511 0.10511 0.10511 0.10511
C22: 1.10492 1.10492 1.10492 1.10492 1.10492

COL 1 2 3 4 5

ROW 3 C11: 1.10511 1.10511 1.10511 1.10511 1.10511
C12: 0.10511 0.10511 0.10511 0.10511 0.10511
C22: 1.10492 1.10492 1.10492 1.10492 1.10492

COL 1 2 3 4 5

ROW 4 C11: 1.10511 1.10511 1.10511 1.10511 1.10511
C12: 0.10511 0.10511 0.10511 0.10511 0.10511
C22: 1.10492 1.10492 1.10492 1.10492 1.10492

COL 1 2 3 45

ROW 5 C11: 1.10511 1.10511 1.10511 1.10511 1.10511
C12: 0.10511 0.10511 0.10511 0.10511 0.10511
C22: 1.10492 1.10492 1.10492 1.10492 1.10492

NOTE: Correct values are C11 = C22 = 1.10500, C12 = 0.10500.


-0












Table 17. Right stretch tensor.


COL 1 2 3 4 5

ROW 1 U11: 1.05005 1.05005 1.05005 1.05005 1.05005
U12: 0.05005 0.05005 0.05005 0.05005 0.05005
U22: 1.04996 1.04996 1.04996 1.04996 1.04996

COL 1 2 3 4 5

ROW 2 U11: 1.05005 1.05005 1.05005 1.05005 1.05005
U12: 0.05005 0.05005 0.05005 0.05005 0.05005
U22: 1.04996 1.04996 1.04996 1.04996 1.04996

COL 1 2 3 4 5

ROW 3 U11: 1.05005 1.05005 1.05005 1.05005 1.05005
U12: 0.05005 0.05005 0.05005 0.05005 0.05005
U22: 1.04996 1.04996 1.04996 1.04996 1.04996

COL 1 2 3 4 5

ROW 4 U11: 1.05005 1.05005 1.05005 1.05005 1.05005
U12: 0.05005 0.05005 0.05005 0.05005 0.05005
U22: 1.04996 1.04996 1.04996 1.04996 1.04996

COL 1 2 3 4 5

ROW 5 U11: 1.05005 1.05005 1.05005 1.05005 1.05005
U12: 0.05005 0.05005 0.05005 0.05005 0.05005
U22: 1.04996 1.04996 1.04996 1.04996 1.04996

NOTE: Correct values are U11 = U22 = 1.05000, U12 = 0.05000.












Table 18. Rotation tensor.


COL 1 2 3 4 5

ROW 1 R11: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011
R12: -0.86596 -0.86596 -0.86596 -0.86596 -0.86596
R21: 0.86596 0.86596 0.86596 0.86596 0.86596
R22: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011

COL 1 2 3 4 5

ROW 2 R11: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011
R12: -0.86596 -0.86596 -0.86596 -0.86596 -0.86596
R21 0.86596 0.86596 0.86596 0.86596 0.86596
R22: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011

COL 1 2 3 4 5

ROW 3 R11: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011
R12: -0.86596 -0.86596 -0.86596 -0.86596 -0.86596
R21: 0.86596 0.86596 0.86596 0.86596 0.86596
R22: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011

COL 1 2 3 4 5

ROW 4 R11: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011
R12: -0.86596 -0.86596 -0.86596 -0.86596 -0.86596
R21: 0.86596 0.86596 0.86596 0.86596 0.86596
R22: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011

COL 1 2 3 4 5

ROW 5 R11: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011
R12: -0.86596 -0.86596 -0.86596 -0.86596 -0.86596
R21: 0.86596 0.86596 0.86596 0.86596 0.86596
R22: -0.50011 -0.50011 -0.50011 -0.50011 -0.50011


NOTE: Correct
0.86603, R22 =


values are R11 = -0.50000, R12 =
-0.50000.


-0.86603, R21 =


m













Table 19. Lagrangian strain from deformation gradient.


COL 1 2 3 4 5

ROW 1 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 2 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 3 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 4 E11: 52554. 52554. 52554. 52554. 52554.
E12: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

COL 1 2 3 4 5

ROW 5 E11: 52554. 52554. 52554. 52554. 52554.
E2: 52557. 52557. 52557. 52557. 52557.
E22: 52459. 52459. 52459. 52459. 52459.

NOTE: Units are micro inches-per-inch. Correct values are Eii =


52,500.


-
















CHAPTER IV
EXPERIMENTS



IV.1 Test Specimen

The prismatic beam shown in Figure 20 was used for all experi-

ments. A clear plastic, CR-39, was chosen for its relatively low

modulus of elasticity and capacity for large, elastic strain.

The 11X11 test pattern on the beam was generated by the synthe-

tic image generator and applied by contact printing. First, a hard

copy of the pattern was photographed using Kodalith film. After

coating the beam with a buffer of polyurethane and naphtha, a Liquid

Light emulsion was applied, contact printed, and chemically fixed

(see Figure 21).




IV.2 Test Equipment

Three experimental set-ups were configured; two were for strain

tests and the third for rigid-body motion.

The strain set-up was a typical photoelastic load frame with

special fixtures (see Figure 22). The fixtures were for four-point

bending (see Figure 23) and cantilever beam (see Figure 24) experi-

ments.

The equipment shown in Figure 25 was used for the rigid-body

motion tests. A photocopy of a protractor was taped to the back of

the lens holder and rotation was measured relative to the tip of the






















S2 Drix '", 11xl Spot Pattern


-~-I--~______-- ____ I~-~ -- -~------c ------i-
S' 1 2" -- 2" ------ 1"J -- ,-
__________________ 2 --------------------3-
7"Figure 20. Beam fabricated from CR-3.













Figure 20. Beam fabricated from CR-39.


I ***

i**






**
J *



0, 0
i^


*****
**.**
*****


*****
O O O
O 0
0
O
O
o-Le


S.'
0* (
00
0@(
0O(
00
00 1
S.(
00


t


i
] '
I
t

i r


* *
l ** .. .. ..


Figure 21. Spot pattern on beam.


4 #28 Drill


- -L-


_~~__~ ~






80


Figure 22. Experimental set-up of load frame.


Figure 23. Set-up for four-point bending.






































Figure 24. Set-up for cantilever beam.


Rigid-body motion set-up.


Figure 25.










clamped pointer. Measurements were accurate to about 1 degree. A

dial indicator measured translation in the horizontal direction with

an accuracy of 0.0005 inch.

Since image processing equipment was unavailable at the

University of Florida, the equipment for the rigid-body experiments

was transported to USC where the digital images were recorded.



IV.3 Experimental Procedure

Unfortunately, the strain tests were not done because the load

frame was too large to transport to USC. The second order model in

equation (53) was for these tests.

The equipment shown in Figure 25 and an image processing system

similar to the one shown in Figure 3 (based on the IBM PC) was used

for the rigid-body experiments. The CCD camera was a Sony model

XCM-38 with a zoom lens, and the digitizer board was an eight-bit

Datacube "Frame Grabber" with 385 rows and 374 columns of pixels.

The set-up included several routine checks. First, alignment

between the rail, table, and camera were checked to minimize experi-

mental error. Then, the zoom lens was adjusted to enlarge the

pattern image as much as possible. Resolution in the horizontal

direction was measured at 358 pixels per inch.

Lighting was adjusted to provide the usual illumination for

tests in the USC lab, and the transparent specimen caused some

lighting problems. Contrast was impaired by the light transmitted

through the beam.





83



After set-up, images were digitized and stored on floppy disk.

All images were transferred to magnetic tape via USC's VAX-11/780

computer for transport to the University of Florida.
















CHAPTER V
ANALYSIS OF RESULTS



The analysis of results has been divided into two sections:

numerical and experimental. The numerical section summarizes synthe-

tic image work and the experimental section summarizes the analysis

of the images digitized at USC.



V.1 Numerical

The analysis of synthetic images included rigid-body motion,

homogeneous strain, combined motion, and the influence of noise, gray

levels, and spot radius. To ensure a good random sample, the results

from a variety of spot spacings were averaged. The pattern was 5X5

with the icon being spot (3,3). Except as noted, the gray level

difference was 200, and the zero-mean noise was Gaussian with a

standard deviation of 7.0 gray levels. The absolute values of noisy

pixels with values less than zero were used instead of the negative

gray levels.

In the following tables, the average absolute error is defined

as follows: average absolute error is the average of the absolute

values of the errors. For example, if E11=E12=E22=0 and the computed

values are E11=100p, E12=-150p, and E22=50p, the average absolute

error is 100p.










Standard deviation, also in the tables, is the standard devia-

tion of the computed values. For the strain example above, the

standard deviation is 132y.

For translations of various amounts in the x and y directions,

the average absolute error was 0.008 pixels with a standard deviation

of 0.011 pixels. Translations near whole pixels had slightly less

error. (Equation (38) suggested a spatial sensitivity.)

Rigid-body rotation was about the image center which was approx-

imately two-thirds of a pixel below the pattern center. Rotations

ranged from 0 to 180 degrees, and the results, computed using the

elements of the rotation tensor, were very consistent with an average

absolute error of 0.0039 degrees and a standard deviation of 0.0008

degrees.

Several strain cases were investigated and summarized in Table

20. Normal strains were E =E22 with E12=E21=0; shear strains were

E =E 22=0 with E 12=E21; and combined strains were E 11=E22=E=E21'

See equation (12) for strain definitions.

Noise had a significant effect on results. As suggested by

equation (38), increasing SNR decreased translation and strain error

(see Table 21). Framing failures began occurring below SNR=100.

Gray level was another significant variable because of the SNR

and the digitization of illumination intensity. For constant noise,

increasing gray level difference increased the SNR and thereby in-

creased accuracy. With a noise standard deviation of 7.00 gray

levels, spots were frequently unidentifiable below a gray level













Summary of strain analysis for synthetic images.


Strain
Displacement
Strain Gradient Ave. Abs. Standard
Case aui/axj Correct Error Deviation


Normal 0.000010 10 17 10
0.000100 100 17 15
0.001000 1001 22 27
0.010000 10050 40 32
0.100000 105000 40 49


Shear 0.000010 10 24 0
0.000100 100 29 0
0.001000 1000 25 0
0.010000 10000 33 0
0.090000a 90000 34 0


Combined 0.000010 10 3 14
0.000100 100 7 28
0.001000 1000 23 26


a Limited by pattern geometry.
NOTE: Rigid-body motion was
per-inch. SNR=816.


zero. Strain units are micro inches-


Table 20.










The effect of SNR on translation and strain calculations.


Translation Strain

Ave. Abs. Standard Ave. Abs. Standard
SNR Error Deviation Error Deviation


100 0.0049 0.0362 39 58
200 0.0029 0.0253 19 42
400 0.0021 0.0176 20 35
600 0.0016 0.0143 14 32
800 0.0012 0.0123 10 26
1000 0.0012 0.0109 8 21
2500 0.0008 0.0068 5 8
5000 0.0006 0.0043 3 6
10000 0.0004 0.0034 2 5
S0.0001 0.0001 0 0

NOTE: Translations were 10.5 pixels in x and y. No rotation. All
strains were Eij=100P.


Table 22. The effect of gray level difference on translation and
strain calculations.



Translation Strain

Ave. Abs. Standard Ave. Abs. Standard
H Error Deviation Error Deviation


32 0.0257 0.0270 ** **
64 0.0043 0.0349 66 62
128 0.0026 0.0172 20 45
256 0.0009 0.0089 3 16
512 0.0007 0.0043 5 8
1024 0.0003 0.0022 4 5
2048 0.0002 0.0011 2 2
4096 0.0001 0.0006 1 2


NOTE: Translations were 10.5 pixels
strains were Eij=100I.


in x and y. No rotation. All


Table 21.









difference of 32. Below a difference of 64, a strain of Eij=100U was

undetectable (see Table 22).

An increase in gray level difference reduced the error caused by

digitization. Without noise, a difference of 16 gray levels produced

an average absolute strain error of 12p (for Eij=100p) with a stand-

ard deviation 0.031 and, above a difference of 64, error was nil.

However, SNR effects greatly overshadowed digitization error.

The effect of spot radius decreased rapidly with increased

radius (see Table 23). Equation (38) suggested this for small 5,

and, in fact, almost no change occurred for radii above 10 pixels.





Table 23. The effect of spot radius on translation and strain
calculations.



Translation Strain

Spot Ave. Abs. Standard Ave. Abs. Standard
Radius Error Deviation Error Deviation


2 0.0030 0.0233 29 20
4 0.0018 0.0272 10 32
6 0.0013 0.0085 7 15
8 0.0016 0.0069 8 13
10 0.0020 0.0052 8 19
12 0.0021 0.0059 5 11

NOTE: Translations were 10.5 pixels in x and y. No rotation. All
strains were Eij=100p. SNR=816.
13


How much VAX 11/750 computer time did the analysis of a 256X256

pixel image require? For a 3X3 spot pattern, CPU time was about 4.9




Full Text
PATTERN MAPPING IN
PLANE MOTION ANALYSIS
BY
R. WALLACE FAIL
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1987

Copyright 1987
by
R. Wallace Fail

To the Glory of God
Digitized by the Internet Archive
in 2010 with funding from
University of Florida, George A. Smathers Libraries with support from Lyrasis and the Sloan Foundation
http://www.archive.org/details/patternmappinginOOfail

ACKNOWLEDGMENTS
I would like to express my appreciation to the members of my
supervisory committee, Dr. S. S. Ballard, Dr. P. Hajela, Dr. U. H.
Kurzweg, Dr. E. K. Walsh, and Dr. C. E. Taylor, chairman.
Support of the graduate program by former department head, Dr.
K. Millsaps, and current department head, Dr. M. A. Eisenberg, is
gratefully acknowledged.
Special thanks go to Mr. Steve McNeil and Mr. Marc Paquette at
the University of South Carolina for their help and extending the use
of their laboratory facilities to digitize the images used in this
study.
I am indebted to my wife, Robyn, for her unwavering support and
love.
IV

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS iv
LIST OF TABLES vii
LIST OF FIGURES ix
ABSTRACT xi
CHAPTERS
I INTRODUCTION 1
1.1 Background 1
1.2 Purpose of Present Work 3
1.3 Scope of Present Work 3
1.4 Survey of Previous Work 5
1.4.1 Solid Mechanics 5
1.4.2 Image Processing/Pattern Recognition 7
II THEORETICAL FOUNDATIONS 8
11.1 Solid Mechanics 8
11.2 Image Processing and Pattern Recognition 14
11.2.1 Digital Image 14
11.2.2 Preprocessing 18
11.2.3 Segmentation 18
11.2.4 Image Features 24
11.2.5 Feature Extraction 26
11.2.6 Known Patterns 39
11.2.7 Syntactic Pattern Recognition 41
11.2.8 Approximation of Syntactic Pattern
Mapping 42
III IMAGE GENERATION AND ANALYSIS 58
111.1 Synthetic Images 58
111.2 Image Analysis 60
v

IV EXPERIMENTS 78
IV.1 Test Specimen 78
IV.2 Test Equipment 78
IV.3 Experimental Procedure 82
V ANALYSIS OF RESULTS 84
V.1 Numerical 84
V.2 Experimental 89
VI CONCLUSIONS AND RECOMMENDATIONS 94
VI. 1 Conclusions 94
VI.2 Recommendations 95
APPENDIX BORDER FOLLOWING ALGORITHM 98
REFERENCES 99
BIOGRAPHICAL SKETCH 104
vi

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
LIST OF TABLES
Page
Unordered spot coordinates in the undeformed image 45
Unordered spot coordinates in the deformed image 46
Ordered spot coordinates in the undeformed image 51
Ordered spot coordinates in the deformed image 52
Spot distance from origin in undeformed image 53
Spot distance from origin in deformed image 54
Spot distance from Icon in undeformed image 55
Spot distance from Icon in deformed image 56
Input file for synthetic image generator 59
Image analysis options file 60
Displacement of each spot 61
Displacement gradients 70
Lagrangian strain from displacement gradients 71
Strain from Taylor series 72
Deformation gradient 73
Green deformation tensor 7 4
Right stretch tensor 75
Rotation tensor 76
Lagrangian strain from deformation gradient 77
Summary of strain analysis for synthetic images 86
The effect of SNR on translation and strain calculations....87
vii

22 The effect of gray level difference on translation and
strain calculations 87
23 The effect of spot radius on translation and strain
calculations 83
24 Images used for rigid-body motion experiments 89
25 Results of rigid-body rotation experiments 90
26 Results of rigid-body translation experiments 92
viii

LIST OF FIGURES
Figure Page
1 The motion of body 8 through space 9
2 The motion of the neighborhood of particle p 10
3 A typical PC-based image processing system 15
4 A spot and its digital image 16
5 A pixel and its eight-neighbors 17
6 The bi-modal histogram of figure 4 19
7 Black and white regions in a bi-modal histogram 21
8 Frame of spot in figure 4 24
9 Image artifacts 25
10 One-dimension model 30
11 The discretization of and kd 33
i D
12 Icon rotation measured from the vertical 35
13 Icon rotation is determined to within a constant by
equation (40) 37
14 Patterns that are easily recognized and analyzed 40
15 Patterns of Figure 14 with Icons 40
16 A pattern and its context-sensitive language 42
17 Image addresses and offsets 43
18 A simple example of motion 44
19 The deformed image and its reference position 48
20 Beam fabricated from CR-39 79
IX

21 Spot pattern on beam 79
22 Experimental set-up of load frame 80
23 Set-up for four-point bending 80
24 Set-up for cantilever beam 81
25 Rigid-body motion set-up 81
x

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PATTERN MAPPING IN
PLANE MOTION ANALYSIS
BY
R. WALLACE FAIL
May 1987
Chairman: C. E. Taylor
Major Department: Engineering Sciences
A new, highly automated method for measuring plane motion,
pattern mapping, has been developed for rigid-body motion and strain
analysis. Pattern mapping employed image processing and syntactic
pattern recognition principles to recognize a known pattern before
and after motion. Using the Lagrangian definition of motion, points
in the two images were mapped, and the map was used to determine
rigid-body motion and strain. Whole-image analysis was thoroughly
demonstrated.
Known patterns were, in this case, rectangular grids of photo¬
dots. These patterns were applied to the specimen by contact print¬
ing. An icon was included in the pattern, thereby accommodating
rigid-body rotations up to one revolution.
Motion functions were selected by the user and easily changed to
suit the particular application. Function coefficients were deter¬
mined by least squares.
xi

Accuracy depended primarily on the signal-to-noise ratio and the
gray level difference between "black" and "white." An increase in
these variables improved accuracy.
User and computer time requirements were quite modest. Images
were processed with little or no user interaction, and output speci¬
fications were automatically read from a disk file. Depending on
output specifications, computer time (VAX 11/750) for an 11X11
pattern in a 384X374 image was 25 to 38 CPU seconds.
xii

CHAPTER I
INTRODUCTION
I.1 Background
The primary purpose in experimental solid mechanics is to deter¬
mine stress and motion (strain, rigid-body translation, and rotation)
from measurements of physical quantities. For example, stress may be
computed from measurements of load and cross-sectional area, and
strain is measured by changes in characteristic lengths. Many
methods (mechanical, electrical, optical, etc.) are available to
analyze stress and motion [1-4].
Optical methods play an important role in experimental mechan¬
ics. They have a number of advantages which include [5,6]
1. Noncontacting measurements do not interfere with material
response.
2. Full-field analysis is usually possible.
3. Highly accurate results are usually possible.
4. Response times are as fast as light, thereby accommodating
dynamic investigations.
5. Actual structures under live loads can frequently be ana¬
lyzed.
6. Effects of actual boundary conditions can be studied in
detail.
1

2
Although optical methods offer many advantages, they do have
several undesirable features. For example, some methods are sensi¬
tive to small extraneous vibrations and require expensive, delicate
equipment. Perhaps the most undesirable feature is the fact that
most of these methods do not directly yield their wealth of informa¬
tion about stress or motion in a readily usable form.
The extraction of this information is time-consuming and re¬
quires special skills. An expert may spend hours reducing data and,
all too often, the amount of data analyzed is limited simply because
automated analysis is not available. Furthermore, results can be
affected by the investigator's particular techniques.
Digital computers are opening doors to automation, increased
accuracy of existing methods, and previously intractable or unknown
techniques. Some of the latest innovations employ digital image
processing and pattern recognition (IP/PR) technologies which utilize
digital computers and video imaging equipment for direct data acqui¬
sition.
The fields of IP/PR are rapidly expanding and are well-
established in many areas. In experimental mechanics, their poten¬
tial for efficiently gathering vast quantities of data consistently
and accurately is being quickly developed. These technologies are
freeing experimentalists from hours of painstaking labor and provid¬
ing them with a new avenue to existing numerical methods and analyti¬
cal tools [7-14] .

3
I.2 Purpose of Present Work
The IP/PR technologies have their price. For example, IP/PR
programs that use fast Fourier transforms or correlation (template
matching) are frequently considered computationally expensive because
they entail large numbers of (relatively slow) floating-point calcu¬
lations. Interactive programs provide great flexibility, but they
need the user's constant attention. Iterative and backtracking
(systematically retracing previous steps) methods can require sub¬
stantial amounts of computer memory for intermediate data storage
[15-18].
Significantly reducing these time and memory requirements is
certainly a desirable goal. Perhaps this is possible by rethinking
the problem and developing a highly automated method which capital¬
izes on the strengths of the digital computer. Thus, the purpose of
this work is to develop such a method, pattern mapping, to obtain
data via digital imaging equipment and subsequently determine plane
motion.
I.3 Scope of Present Work
Although it was not apparent in the beginning, the concepts
developed in this investigation are readily applied to the photodot
(grid) method [19]. The scope includes this photodot adaptation.
Actually, this investigation encompasses a variety of scientific
disciplines. Perhaps the best approach is to consider the attributes
of each and work toward a goal that is obtainable in a reasonable
amount of time. Specific objectives included in the scope are to

4
1 . guarantee unique mapping
2. provide subpixel registration of 0.1 pixel
3. provide a wide range of motion measurement
4. provide whole-image analysis
5. reduce computer and user time as much as possible
6. provide highly automated analysis.
(Pixels are discrete elements in a digital image, and whole-image
implies results are computed for every photodot.) Special emphasis
is on accuracy and increased computation efficiency.
Several laboratory experiments using a CR-39 (plastic) prismatic
beam were included in the study. The experiments were translation,
rotation, general rigid-body motion, four-point bending, and
cantilever beam. All experiments were static cases of motion in a
plane normal to the optical axis. A rectangular (11X11) pattern of
photodots was contact printed on the lateral side of the beam.
For convenience, motion is defined only in terms of the
Lagrangian formulation. Evaluation of motion functions are limited
to a linear and a second order approximation. The least squares
method of curve fitting is used exclusively, and normal equations are
solved by matrix inversion using Gauss-Jordan elimination.
Synthetic images are used to debug analysis programs and inves¬
tigate the influence of various factors. Motion in the images in¬
clude affine transformations and rigid-body rotation about the image
center. Interpolation of between-pixel intensities is bilinear, and
system-induced distortions are assumed eliminated during preprocess¬
ing. Noise has a zero mean and selectable Gaussian or uniform

5
distributions. Image size is 256X256 pixels which are unsigned, 16
bits or less. The camera position is assumed fixed.
Segmentation and feature extraction (dividing the image into
subregions and obtaining useful information) include neither prefil¬
tering nor histogramming (gray levels) and are based on the random
selection of a few pixels. Processing is automatic and direct
because known patterns in the image are always completely visible and
recognizable. Feature space includes gray levels, areas, centroids,
and second moments. The pattern language consists of only one sen¬
tence, and it is context-sensitive.
1.4 Survey of Previous Work
1.4.1 Solid Mechanics
Early applications of image processing in experimental mechanics
involved the analyses of fringes. The fringes were from photoelasti¬
city, moire, or holographic interferometry. Burger [7] and Chen [8]
describe several of these applications in detail.
Another image processing technique applied to mechanics is cor¬
relation. Researchers [9,20-24] determine plane displacements (uif
i = 1,2) and their gradients Qu^/OXj, i,j = 1,2) by comparing digitally
recorded images of the same random pattern. For example, a pattern
of black speckles (e.g., minute specks of black paint) on a white
field is recorded as integer gray levels at discrete points
(pixels). Two images are used in the analysis: one before and one
after motion. By comparing (spatially registering) small subimages
from each, displacements and their gradients are determined at

6
selected points in the (original) pattern. Sigler and Haworth [25]
combine holography with correlation to determine motion. Thus, the
correlation method has the advantages of using coherent or
conventional light sources and easily applied patterns.
The grid method is one of the oldest and simplest methods used
in the analysis of strain. One simply applies reference marks
(lines, spots, scratches, etc.) to the surface of a specimen and
records the original distances between marks. Next, the specimen is
deformed and the distances are again recorded. Then, changes in the
distances are determined and strain is computed [1,2,26-32].
Although simple, the grid method is tedious and traditionally
limited to larger strains. Typical strain measurements are greater
than 0.001 inches-per-inch; the results depend on the care exercised
and the instrumentation [19,26,28]. Even when optical comparators
are used for the smaller strains, many man-hours are frequently
needed to complete the analysis of large grids.
Examining a specimen through a comparator and determining the
position of a grid mark is also a subjective process. Some investi¬
gators define a mark's position by its left edge, some by its right
edge, and some by a variety of definitions. All these definitions
are subject to error caused by rotation, the nonlinear characteris¬
tics of the recording medium (if photographically recorded), grid
application techniques, lighting, operator fatigue, etc. [19].
Parks [32] suggested computer technology might improve the
accuracy of the method and would certainly reduce the labor

7
involved. Previous efforts to fully computerize the grid method have
not been reported in the literature.
Sevenhuijesen [33.3^] reported some feasibility studies with a
Reticon photo-diode array camera. Using an eight-bit digitizer and
digitizing the signal from each diode, he obtained a resolution of a
few hundred micro inches-per-inch. Images were recorded photographi¬
cally without the aid of an image processing system.
1.4.2 Image Processing/Pattern Recognition
Several researchers active in IP/PR have been interested in two-
and three-dimensional motion analysis. Most efforts have concentrat¬
ed on the motion of rigid bodies, clouds, the human body, optical
flow, etc. Motion parameters were obtained by a variety of iterative
processes [35-38].
Huang and Tsai [39-43] proposed a direct method of mapping the
three-dimensional motion of a plane using two, time-sequential
images. Mapping parameters were determined uniquely by solving a set
of linear equations. Unfortunately, motion parallel to the optical
axis could not be determined, and "small" approximations were made
for strain and rotation. Mathematical proofs guaranteeing unique
mapping between images were included.

CHAPTER II
THEORETICAL FOUNDATIONS
II. 1 Solid Mechanics
The description of motion used here is limited to the Lagrangian
description and is consistent with the classical field theories of
modern continuum mechanics [44-48]. The Lagrangian description is
sufficient to determine the directions of the principal axes of
strain and the magnitudes of the principal stretches. It is, there¬
fore, a fully general measure of strain and is capable of describing
small or large strains. For convenience, all reference frames are
Cartesian in three-dimensional Euclidean space.
In Figure 1, body 8 consists of a set of particles which assume
a continuous progression of configurations in time. The undeformed
configuration is chosen as the reference state where particle P
occupies the "material" position X at time t=tg. Subsequent motion
carries particle P to the "spatial" position x in the deformed or
"after motion" state at time t. Although each state may be defined
in terms of its own coordinate system, material and spatial coordi¬
nates are usually measured with respect to the same coordinate
axis. The motion of a point is described by the relation of the
coordinates between the two states. Mathematically, motion is ex¬
pressed by a one-to-one mapping x symbolized as
8

9
x = x(X,t)
or x.
i
X. (1)
B
Figure 1. The motion of body g through space.
The deformation gradient of the motion x is defined as the
derivative or gradient of x at X and is denoted by
F
3x
ax
or
F. .
ij
(2)
where F is the fundamental quantity for the analysis of local proper¬
ties of deformation, and the physical components are, of course,
dimensionless as are all measures of relative configuration.

10
Alternatively, motion written in terms of the displacement
vector u is
x = X + u(X,t)
or x.
i
X.
i
+ u.(Xi,X2,X ,t)
(3)
And the derivative of u, the displacement gradient, is
J
3u
3X
or
au.
i
ax.
j
(4)
If two particles in the reference configuration are an infini¬
tesimal distance dX apart, then these two particles are an infinites¬
imal distance dx apart in the deformed configuration (see Figure
2). Tensor F maps the neighborhood of particle P in the
Figure 2. The motion of the neighborhood of particle P.

reference configuration to its neighborhood in the deformed configur¬
ation. Thus, the linear transformation is
dx = F*dX or dx. = F. .dX.
~ ~ i ij j
(5)
Unique mapping is guaranteed as long as the Jacobian J = |F| is
greater than zero and finite.
Since vector dX has length dS and vector dx has length ds,
Lagrangian strain E may be defined in terms of these lengths squared
as
(ds)2 - (dS)2 = 2dX-E-dX
(6)
The Green deformation tensor C, referred to the undeformed con-
figuration, gives the new squared length (ds) of the element into
which the given element dX is deformed. Symbolically, the relation¬
ship is
(7)
where
T
C = F -F
(8)
Conversely, the Cauchy deformation tensor B 1 gives the initial
2
squared length (dS) of an
element dx in the deformed

12
configuration. This relationship is written as
(dS)2 = dx-B 1 *dx (9)
where
-1 -1 T -1
B = (F ) -F (10)
Lagrangian strain written in terms of the Green deformation
tensor is
(11)
In terms of the displacement gradient, E is
9u. 9u. 9u 9u
E . = -[—- + —¿ + —- —-
ij 2L9X. 9X. 9X. 9X.
J i i J
(12)
Strain, computed by a Taylor series [1], is
E
11
J
1+2— + ( —)2+(^X)2+(l-)2
9X V9X; V9X; V9X;
-1
E
22
4
1+2— +(—)2+(—)2+(—)2
9Y v 9Y' V9Y; V9YJ
(13)
E
12
"21
— arcsin[
9u
9Y
9v | 9u 9u |
9X + 9X 9Y
9v 9v
9X 9Y +
9w 9w
9X 9Yj
2>

13
For many engineering applications, the so-called "small dis¬
placement" theory is adequate and much easier to implement. By this
theory, strain is
3u
£11 3X
1 r3u 3v,
e = —r— + —i
12 2l3Y 3Xj
(14)
3v
£22 “ 3X
If the Jacobian J is greater than zero and finite, the deforma¬
tion gradient written in terms of a rotation tensor R and the right
stretch tensor U, a polar decomposition, is
F = R-U (15)
Tensors R and U are unique and represent rigid-body rotation and
stretch, respectively. Rotation tensor R is orthogonal and produces
a rigid-body rotation between the principal axes of C at X and the
principal axes of B 1 at x. The necessary and sufficient condition
for no local rigid-body rotation (sometimes called "pure strain")
during motion is R=1 . Symmetric, positive-definite U produces the
changes in vector lengths during motion and produces rotation, in
addition to R, of all vectors except those in the principal

14
directions of U. Note that U=1 is necessary and sufficient to define
locally "pure" rigid-body motion (neglecting the trivial case when no
motion occurs), and translation does not change a vector with respect
to the common reference axis. Tensors U and R, written in terms of
F and C, are
U
= C
1/2
(16)
R = F*U
(17)
As a final note, the deformation at any point may result from
translation, a rigid-body rotation of the principal axes of strain,
and stretches along these axes. These motions may occur in any
order, but their tensorial measures may not. Mathematically,
Lagrangian motion is assumed as a successive application of
1 . stretch by U
2. rigid-body rotation by R
3. translation to x.
II.2 Image Processing and Pattern Recognition
II.2.1 Digital Images
A typical image processing system is shown in Figure 3. Conver¬
sion of the intensity distribution of a camera image to a form suit¬
able for computers, digitizing, is accomplished by the digitizer.
After digitization, the image is transferred to the

15
computer for further processing. How an image is processed depends,
of course, on the hardware configuration [15,18,35,49-51].
Object Microcomputer Digitizer
with Internal Monitor
Digitizer
Figure 3- A typical PC-based image processing system.
A digital image is an approximation of a real, continuous
image. For example, the real image of the spot in Figure 4a is re¬
presented in digital form as the matrix of integer numbers in Figure
4b. Each matrix element (or pixel which is dimensionless) corre¬
sponds to a discrete point in the real image, and the value of the
element (gray level) is the approximation of the average local illum¬
ination intensity.
Notice that black has lower gray levels than

16
white and that the edge region (boundary) between the two has inter¬
mediate values. The variation of gray levels in regions of constant
illumination is the effect of noise. Gray levels range from zero to
some maximum which depends on the hardware (e.g., camera, digitizer).
H9
151
155
146
163
158
158
151
159
162
151
1*2
145
146
156
1*5
160
148
165
157
145
156
142
146
159
127
148
158
166
151
153
142
1*8
137
149
133
148
143
15»
155
156
145
146
159
159
166
156
152
154
153
155
132
148
144
151
169
164
1*4
165
156
138
155
1*6
1*8
129
196
98
115
127
159
154
137
149
152
151
153
15»
151
147
156
148
133
83
47
36
51
56
69
99
142
134
147
163
146
159
1*6
152
146
134
138
66
*5
52
48
49
47
51
79
52
139
152
150
150
146
1*5
154
140
141
84
56
56
47
59
43
51
46
53
59
91
151
153
153
151
1*6
155
143
138
53
61
51
50
55
*3
48
51
*5
56
44
135
154
146
154
15»
159
150
100
51
*8
36
44
55
49
51
58
53
59
49
195
141
144
159
153
135
152
96
50
59
44
52
43
56
57
5»
5*
46
47
191
150
143
143
156
155
155
108
40
*3
46
49
59
41
48
49
55
51
42
193
146
158
152
153
140
166
137
47
51
47
41
48
37
51
5»
59
47
33
137
156
153
143
1*7
149
151
153
85
*5
59
55
49
5»
51
*8
44
43
82
165
147
153
169
1*7
135
155
159
138
76
55
43
57
53
54
*7
59
63
132
151
138
147
148
1*5
1*8
151
162
159
1*4
92
50
55
59
53
39
79
141
148
152
148
159
152
151
153
156
153
148
132
1*8
132
199
96
99
1*3
154
159
145
154
163
159
151
1*4
141
149
157
148
156
145
138
145
154
142
139
145
145
147
154
161
148
149
157
159
141
144
158
1*9
166
150
151
142
142
159
145
156
150
159
151
147
149
156
158
146
160
148
1*9
148
149
151
144
149
149
159
151
153
152
149
157
169
155
147
150
131
150
159
163
150
158
153
147
1*6
163
136
151
154
153
152
153
151
169
146
148
144
152
153
148
169
169
136
157
1*4
144
141
162
146
148
154
Real laage
a)
Digital Iaage
b)
Figure 4. A spot and its digital image. The values, I(x,y), in the
digital image correspond to illumination intensity in the
real image.
Noise corrupts the image gray levels, and the sources may be
classified into two groups: optical and electronic. Two examples of
optical noise are particulate matter in the air and laser speckle.
Electronic noise comes from the discretization process, thermal

17
sources in the hardware, interference from extraneous electromagnetic
sources, etc. Noise is usually assumed statistically random and
normally distributed with a zero mean, p, and a standard devia¬
tion, a . Unfortunately, no known method completely eliminates the
n
effects of noise.
Each array element has neighbors which are commonly called the
eight-neighbors [52]. In Figure 5, the center element, pixel (x,y),
is the reference element or element of interest. The element
directly above it is arbitrarily labeled neighbor 0, and the other
elements are named in a counterclockwise direction. Although ele¬
ments along the image's outer array edge do not have all
1
0
7
2
Pixel
(x.y)
6
3
4
5
Figure 5. A pixel and its eight-neighbors.

18
8 neighbors, the neighbors they do have are also labeled according to
Figure 5.
Spatial and illumination sampling must be sufficient to repre¬
sent adequately the real image. Too low a spatial sampling frequency
averages intensity over too large an area (see Figure 2.7 in refer¬
ence 15), while too high a sampling frequency results in the loss of
information outside the field of view. In general, increasing inten¬
sity corresponds to increased gray levels, but intensities beyond the
capability of the equipment are limited to the maximum gray level.
An increase in the range of gray levels increases image detail (see
Figure 2.8 in reference 15).
11.2.2 Preprocessing
After digitization, the digital image is corrected for geometric
and radiometric distortion. Distortion, introduced by the processing
system, is removed by using the appropriate calibration procedure
[49,531. This type of correction is frequently called preprocessing
and images in this study are assumed corrected unless otherwise
stated.
11.2.3 Segmentation
The next step in image processing, segmentation, consists of
identifying regions and boundaries of interest in the image
[18,54], Of course, the difficulty in segmenting an image depends on
its complexity. Since the images developed in this study are rela¬
tively simple, they are easily segmented.
Histogramming. Segmentation usually begins with histogram-
ming. When the frequency distribution of the gray levels in an image

19
is computed for the entire image, contrast and the presence of multi¬
ple modes are readily observed. Figure 6, the bi-modal histogram of
Figure 4, has an average black of 50 and an average white of 150.
Note, the histogram yields no direct information about the location
of the gray levels in the image [55].
Gray Levels
Figure 6. The bi-modal histogram of Figure 4.
Usually, computing the histogram for the entire image is not
considered computationally expensive. This is not true in the
present example because the percent of the total computer time
devoted to histogramming the entire image is significant. Since the
images used here are designed to have histograms similar to Figure 6,

20
the needed information is easily estimated without actually computing
a histogram.
Thresholding. Thresholding follows histograraming and includes
the task of determining the gray level which, in this case, reason¬
ably separates regions of black from white. This is frequently
accomplished interactively or by use of a priori information, and the
choice of the optimum threshold is a subject of considerable interest
[56-62].
The alternative approach in this study requires neither inter¬
active thresholding nor an optimum threshold. Instead, a good esti¬
mate of gray levels that are "definitely black" and an estimate of
the average white level are used. The estimates are derived from
histogram statistics and based on the following assumption: the
histogram is strongly bi-modal. In Figure 6, gray level 53 is
assumed black and the average white gray level is about 151 . The
task of estimating these levels is a four-step process.
First, pixels are randomly selected and their average gray level
1^ is computed along with the standard deviation o^. A sample size
of 30 or more is statistically sufficient to estimate the mean and
standard deviation of the entire histogram [63,64]. Obviously,
darker pixels are to the left of the average and lighter pixels are
to the right (see Figure 7).
In the second step, the standard deviation is subtracted from
the average, resulting in Ig. Algebraically, this is
(18)

21
Cray Levels
Figure 7. Black and white regions in a bi-modal histogram (1^=132,
I3=98, I=53. I2=151, It=102).
Another set of 30 or more pixels whose values are below Ig are ran¬
domly selected and used to compute the average 1^. Thus, the average
black level is approximated by 1^.
Chebychev's theorem [63] states that, if a probability distribu¬
tion has a mean p and a standard deviation a, the probability of
obtaining a value x which deviates from the mean by at least K stand¬
ard deviations is at most . Mathematically, this is expressed as
¡C
> Ko) < ^
K
P(|x-p
(19)

22
Since only one standard deviation is subtracted from the average,
finding pixels darker than Ig is highly probable. A skew of the
histogram to the left further improves the odds of finding pixels
less than Ig.
The third step is the initial estimation of the average white
gray level, I2. A sufficient random sample whose values are greater
than IA is used to compute the initial estimate of I2.
Finally, the threshold value It is computed using the estimates
for the black and white levels. Symbolically,
Images designed for this study are assumed to have two
regions: black and white. In a sense, they are binary (region)
images. Thus, segmentation is easily accomplished by locating and
isolating the black regions of interest from the white background.
Raster scanning and border following. Beginning at the origin,
the upper left corner, the image is raster scanned (i.e., along a row
or column) until a black pixel is found. If at least one of the
neighbors is black, a possible black region is assumed--spurious
noise is assumed if the pixel has no black neighbors. If a black
neighbor is found, the border between black and white is followed
around the black region thereby isolating it from the neighboring
white.

23
Rosenfeld [18] has developed a general-purpose border-following
algorithm. First, gray levels are reduced by thresholding to 0's for
black or 1's for white. (This is quickly done if one has the
appropriate hardware.) By comparing neighbors and locating the next
edge pixel, the algorithm follows the border counterclockwise for an
outer border and clockwise for an inner one. Since borders may
overlap, provisions are made for backtracking.
In this study, the generality of Rosenferd's algorithm is un¬
necessary, and the algorithm is modified for faster execution. The
modified algorithm (see appendix) is designed to follow outer edges
(no inner edges exist) in the clockwise direction, and reducing the
image is eliminated by combining thresholding with border following.
The only information saved during border following is the y
coordinate of the spot's left-most pixel, L; the y coordinate of the
right-most pixel, R; the x coordinate of the highest pixel, T; and
the x coordinate of the lowest pixel, B. Since the black region
extends a little beyond the approximate border, 2 pixels are sub¬
tracted from L and T, while 2 pixels are added to the R and B (see
Figure 8). Thus, the local region of interest (black and inter¬
mediate gray) is "framed" for further processing.
Raster scanning resumes at the bottom of the frame and continues
until another black region is found and it is framed. Scanning and
framing continue until the entire image is covered. Thus, the image
is completely segmented and the information needed for feature ex¬
traction is available.

24
rY
X
Edge
T
—
Frame
149
151
155
146
163
158
158
151
150
'162
151
142'
145
146
156
145
160
148
165
157
145
156
142
146
150
127
148
158
166
151
153
142
148
157
149
153
148
143
150
155
156
145
146
150
159
166
156
152
154
153
155
132
148
144
151
169
164
144
165
156
138
155
146
148
129
106
98
115
127
150
154
137
149
152
151
153
150
151
147
156
148
138
83
47
36
51
56
60
90
142
154
147
163
146
150 '
146
152
146
134
138
66
45
52
48
49
47
51
70
52
139
152
150
150
146
145
154
140
141
84
56
56
47
50
43
51
46
53
59
91
151
153
153
151
146
159
143
138
53
61
51
50
55
43
48
51
45
56
44
135
154
146
154
150
150
150
100
51
48
36
44
55
49
51
58
53
50
49
105
141
144
150
153
139
152
96
50
50
44
52
48
56
57
50
54
46
47
101
150
143
143
156
155
155
108
40
43
46
49
59
41
48
49
55
51
42
103
146
158
152
L
15?
148
166
137
47
51
47
41
48
37
51
50
50
47
33
137
156
153
Ü3
147
148
151
153
85
45
50
55
49
50
51
48
44
43
82
165
147
153
160
147
135
155
159
138
76
55
45
57
53
54
47
50
63
132
151
158
147
148
145
148
151
162
159
144
82
50
55
56
53
39
70
141
148
152
148
159
152
151
153
156
153
148
132
148
132
109
96
99
143
154
150
145
154
163
159
151
144
141
149
157
148
156
145
158
145
154
142
139
145
145
147
154
161
148
149
157
150
141
144
158
149
166
150
151
142
142
150
145
156
150
150
151
147
149
156
158
146
160
148
140
148
149
151
144
149
149
150
151
153
152
149
157
160
155
147
ISO
131
150
150
163
150
158
153
147
146
163
136
151
154
153
152
153
151
160
146
148
144
152
153
148
160
160
136
157
144
144
141
162
146
148
154
Edge
B
Figure 8. Frame of spot in Figure 4.
11.2.4 Image Features
Images contain various types of information known as features,
and, when properly selected, they adequately represent the images.
Here, it is assumed that natural features are intrinsic to the real
scene and artifacts are added as desired to aid subsequent proces¬
sing. Features, both natural and artifact, are generally grouped
into three classes: physical, structural, and mathematical [65].

25
Physical and structural features are, in this case, the arti¬
facts shown in Figure 9 plus the pattern they form in the image. For
simplicity, the (single) icon is considered a "special" spot, and all
other spots are assumed uniform in size.
Spot
Icon
Figure 9. Image artifacts. The icon is a "special" spot.
Mathematical features are frame areas, spot centroids, and those
concerning icon identification and rotation. These features allow
extensive integer programming which speeds the majority of the cal¬
culations and improves accuracy.

26
II.2.5 Feature Extraction
Here, feature extraction consists of isolating the artifacts and
computing their desired mathematical features. Since the artifacts
are extracted (by framing) during segmentation, all that remains is
the computation.
Frame area. Frame (integer) area is easily computed using the
information obtained during the segmentation process. The computa¬
tion is
A = (R-L)*(B-T) (21)
Since the icon's frame area is much larger than the other spots',
classifying each artifact as "spot" or "icon" is a simple, binary
task.
Spot centroids. The centroids of all spots are used as defini¬
tions of spot position. If the gray levels in a spot's frame are
assumed to be "weights," the x and y coordinates of the centroid are
calculated [18] by
B R
Z Z xl(x,y)
y _ X=T y=L
B R
Z Z I(x,y)
x=T y=L
(22)
R B
Z Z yl(x,y)
y , y=L X=T
R B
Z Z I(x,y)
y=L x=T

27
Henceforth, these summation limits are implied unless otherwise
noted.
How do gray levels, noise, and spot size affect spatial resolu¬
tion of the centroid? Here, it is evident (see Figure 8) that merely
summing with gray levels I(x,y) does not yield the proper result
because the white pixels shift the spot's computed centroid away from
the correct value. A transformation of all gray levels in the frame
by the equation
I'(x.y) = I -I(x,y), (23)
w
where I is the local average white level and I'(x,y) is the noisy
gray level difference (between black and white), gives the approxi¬
mate centroid coordinates. And since pixels on the frame edge are
white, their average gives a good, local estimate for Iw. The sub¬
stitution of I'(x,y) into equations (22) leads to
Z Z x I'(x,y)
V = 2LJL
z Z I'(x,y)
x y
(24)
Z Z y I'(x,y)
y = y x
I £ I’(x,y)
y x

28
To account for noise, assume each gray level in the image con¬
sists of the actual (i.e., uncorrupted by noise) value, r(x,y), plus
zero-mean Gaussian noise, n(x,y). Thus,
I'(x,y) = I - r(x,y) + n(x,y)
w
I'(x,y) = H(x,y) + n(x,y) (25)
where H(x,y) = I - T(x,y) is the actual gray level difference,
w
Substituting equation (25) into equations (24) leads to
EExH + EExn
- = xy xy
EEH + EEn
xy xy
(26)
EEyH + EEyn
y = yx Yx
EEH + EEn
yx yx
If the expected value [63] for the noise is assumed
E{n(x,y)} = EEn(x,y)
xy
= EEn(x.y)
yx
(27)
= En(x,y)
x
= En(x,y)
y

29
and substituted into equations (26), then the expected centroid is
E{X}
e{x}
e{y}
EExH + ExEq
xy
X y
EEH
+ EEq
xy
xy
EExH + E{n}Ex
xy x
EEH + E{n}
xy
EEyH + E{n}Ey
yx y
EEH + Sin!
yx
(28)
Since E{n}=0, equations (28) become
E{X}
EExH
xy
EEH-
xy
E|Y}
EEyH
yx
EEH
yx
(29)
In general, the summations in equations (27) are not equal to
zero, which introduces random error. But if Eq = 0, then X = E{X},
y
and if Eq = 0, then Y = E{Y}. Thus, the summation process used to
x
compute the centroids greatly reduces, although in general it does
not eliminate, the effect of noise.

30
To understand better how noise, spot diameter, and number of
gray levels interact to affect spatial resolution, consider the one-
dimension model in Figure 10. This model is essentially a column in
the two-dimensional image (see Figure 8) with frame edges at x=T and
x=B. For simplicity, edge pixels are always located at x=T+2 and
x=B-2. (The extension to the general two-dimension case is straight¬
forward.) Similar to equation (26), the centroid calculation is
ZxH(x) + Exq(x)
EH(x) + En(x)
X X
(30)
Frame
Figure 10. One-dimension model.
Assume

31
x=T+2
x=B-2
T+3 otherwise
1 - c— m
0< K_<1
T-
°— and T<6 where Hc, a constant over the line length, is the gray level differ¬
ence and Hm is the maximum possible gray level difference (contrast)
between black and white.
The influence of noise and the edge pixels can be described in
terms of resultants. The noise resultant, 6nc, is the product of
random variables 6 and n , i.e., Exn(x)=6ri , and En(x)=n • Dimen-
c c c
X X
sionless parameter is the fraction of the x=T+2 edge pixel covered
by the line; and k H is the resultant. Assume the same for k
I C D
and < H at edge x=B-2. During digitization, k_ and kd are discre-
DC ID
tized (see Figure 11) to d and <1, respectively. For a given line
i D
segment at a specified location, < and I D
stitution of these values into equation (30) yields
B-3
E xH +H [k (T+2)+<(B-2)]+ón
- _ x=T+3 c c T B °
x=iH(X)+Hc(VKB)+T1c
B-3
E x = (T+3)+(T+4)+(T+5)+ ... +(B-5)+(B-4)+(B-3)
x=T+3
KTH0
k0H
B c
H(x) =
If

32
is added to
3-3
E x = (B-3)+(B-4)+(B-5)+ ... +(T+5)+(T+4)+(T+3)
x=T+3
then
B-3
2 E x = (T+B)+(T+B)+(T+B)+ ... +(T+B)+(T+B)+(T+B)
x=T+3
and
B-3
E x = |(T+B)
x=T+3
where n is the number of terms in the sum. Since B-3_>T+3 and n>1 ,
n = (B-3) - (T+3) + 1
and
n = B-T-5.
Therefore,
B-3
E x
x=T+3
(B-T-5)(T+B)
2
B-3
E H(x) = (B-T-5)H
x=T+3
and

33
Subsequently, equation (30) becomes
(B-T-5)(T*B)Hc*2|Snc»HcUT(T*2)»c3(B-2):])
2[ After dividing the top and bottom by H , equation (31) becomes
(B-T~5)(T+B)+2[5 ^+ 2[(B-T-5)+C+ (32)
G
where £ = — . Note, the £ ratio is similar to the inverse of
h
C
Pratt's [66] signal-to-noise ratio,
H 2
SNR = (—)
a
(33)
Fraction of Edge Pixel covered
Figure 11. The discretization of and k . An increase in Hc im¬
proves spatial resolution of the edges.

34
After studying Figure 10 and equation (32), it
that 6=T or 5=B gives the most noise error while
(B-T-5)(T+B)+2[k(T+2)+k(B-2)]
x _ i E
2[(B-T-5)+ gives none.
If a line length of d=B-T~5 is assumed, equation (32)
(d)(d+2T+5)+2[6^+ v = L 5
2[d+5+ic +< ]
If 5=0,
(d)(d+2T+5)+2[kt(T+2)+<_(d+T+3)]
X = 1 £
2[d+ which is the "exact" centroid. If T=0 is assumed and k,
substituted into equation (35), the resolution error is
(d)(d+5)+2C2 Re 2[d+K^,+Kg] 2[d+C+<|+K^]
The general relationship between £, d, and Rg is
equation (37) is simplified. To assume the line position
variable implies the following expected values:
e{kt} = E{ is observed
(34)
becomes
(35)
(36)
' and <' are
r d
3)1
— (37)
clearer if
is a random
After some algebra, equation (37) becomes

35
_ £(d+5-2ó)
e 2(d+5+1)
(38)
For £ much smaller than d, equation (38) suggests an almost linear
relation between 5 and Rg. A spatial dependency (perhaps difficult
to verify experimentally) is also indicated.
Figure 12. Icon rotation measured from the vertical.
Icon rotation. Next, the icon's angle of rotation is computed
(see Figure 12) using the second moment of inertia. The analysis is
as follows:

36
I = EE I' (x,y) d2
xx y
xy
I = Zl I'(x,y) d2
yy „„ x
(39)
P = ZZ I'(x,y) d d
Xy xy X y
where Ixx = moment of inertia about icon's x-axis
I = moment of inertia about icon's y-axis
P = product of inertia about icon's centroid
* y
dx = x distance from pixel to icon centroid
dy = y distance from pixel to icon centroid.
Icon rotation is
2P
e - ilan'1 (J-)
yy xx
(40)
The above equation gives the angle to within a constant. For
example, the analysis gives the same angle for both icon orientations
in Figure 13* The difference is resolved by constructing line "A-A"
through the icon centroid and computing the centroids of the upper
and lower segments. Line "A-A" has the form
y. = mx +b
J A A
where x^ and y^ are points on "A-A" with
m = -TAN(e)
and
b = yT -mx
Icon Icon

37
Y
Figure 13. Icon rotation is determined to within a constant by equa¬
tion (40).
The centroid for the upper segment is
XA
Z
x=T
yA
Z
y=L
xl’(x.y)
XA
yA
Z
Z
I'(x,y)
x=T
y=L
XA
yA
Z
Z yl’(x.y)
x=T
y=L
XA
yA
Z
Z I'(x,y)
x=T
y=L
(41 )

38
and for the lower segment,
X, =
3 R
E E xl' (x, y)
x=xA y=yA
L B R
E E I'(x,y)
x=xA y=yA
(42)
Yr =
B R
E E yl'(x.y)
X=XA y=yA
L B R
E E I’(x,y)
X=XA y=yA
The distances from the icon centroid to the segment centroids are
V(XU XIcon) +(YU YIcon)
(43)
a/(X. -Xt )2+(Y-Yt )2
^ L Icon L Icon
If dn>d. , 9 computed by equation (40) is correct; otherwise 0 must be
U L
shifted by
0 =
f0-ir
0>O
l_0 + ir
0 (44)
Thus, a maximum rotation of one revolution is measurable.

39
Information saved during feature extraction is, of course, frame
area, the centroid coordinates of each spot, the number of spots,
which spot is the icon, and icon rotation. The areas and the cen¬
troids are stored as linear arrays in the order the spots were
found. Lastly, the problem of sorting this raw information and using
it to match spots in the before- and after-motion images is solved
using pattern recognition principles.
II.2.6 Known Patterns
In this study, patterns are classified as either "known" or
"unknown." Known patterns contain structural features (natural,
artifact, or both) that are recognizable before and after motion.
Unknown patterns are patterns which never have recognizable structure
or those in which features are obscured beyond recognition. Known
patterns containing only artifacts are considered here.
Artifacts and their arrangement are selected to improve segmen¬
tation, feature extraction, mapping, and subsequent numerical
analysis. Examples of simple, easily analyzed patterns are shown in
Figure 14. These patterns are fine for small rotations and are
easily recognized because the relative order of the spots does not
change, e.g., the upper left spot remains in the same position--
relative to the other spots—during motion. If rotation is 45
degrees or more for the square patterns and 90 degrees or more for
the others, the relative order is lost. The patterns in Figure 15
allow spot identification after rotations up to one revolution. Once
the patterns in the undeformed and deformed images have been

40
Figure 14. Patterns that are easily recognized and analyzed. Frames
are for reference and not part of the pattern.
Figure 15. Patterns of Figure 14 with icons. The icons are located
for convenience.

41
recognized, subsequent analysis is simplified (especially for the
Lagrangian description) because the undeformed spacing is approxi¬
mately constant.
II.2.7 Syntactic Pattern Recognition
Pattern recognition is usually divided into two classes: deci¬
sion-theoretic and syntactic. In the first class, a pattern is
represented by feature vectors and recognized by a stochastic parti¬
tioning of feature space [67—69]. Syntactic recognition uses pattern
structure and linguistic rules for analysis [70,71].
Syntactic recognition assumes a pattern is described by a
language which contains valid sentences composed from a finite alpha¬
bet or set of symbols. The language is governed by "pattern grammar"
or set of grammar rules. In context-free languages, pattern grammar
is unrestricted. Context-sensitive languages restrict pattern
grammar on the number of sentences, sentence construction, or both.
Here, the pattern language is context-sensitive and has only one
sentence which is composed of a fixed number of ordered symbols. The
alphabet symbols are the spots' relative positions whose order is
determined in the undeformed image. A spot's context is determined
by its 0,2,4,6 (or alternatively by 0 through 7) neighbors. For
example, (2,2) has neighbors (1,2), (2,1), (3,2), and (2,3). An
attempt to map (2,2) in the undeformed image to, say (1,1), in the
deformed is "out of context" with the neighbors (see Figure 16).
Thus, correct pattern mapping is assured if every spot maps in con¬
text with its neighbors.

42
<1.
1)
(X.
2)
(1.
3 >
(1.
4)
(1.
5)
<2,
1 )
(2.
2)
(2.
3)
(2.
4)
<2.
5)
(3.
1 )
(3.
2)
(3.
3)
<3.
4)
(3.
5)
<4,
1)
(4.
2)
(4,
3)
(4.
4 >
(4.
5)
(5,
1)
<3.
2)
(3,
3)
(5,
4)
(3.
3)
Pattern
One-sentence Language
Figure 16. A pattern and its context-sensitive language. Note the
alphabet (symbols) and context in the language.
Higher dimensional pattern grammars [70~72] provide the capabil¬
ity to formally describe, recognize, and map patterns similar to
those in Figure 15. Since a record of each spot and its neighbors
must be maintained and manipulated, generating and parsing formal
grammar for large patterns becomes an intense process. If possible,
trading some of the formal elegance for an economic approximation is
highly desirable.
II.2.8 Approximation of Syntactic Pattern Mapping
Fortunately for many motions, spot relative positions change
very little, which allows context to be determined from spot

43
"addresses." If both images are carefully divided into a number of
addresses (subregions), then a spot has the same address before and
after motion--or nearly so. The determination of a spot's address is
a simple calculation involving centroid coordinates, and recognition
is reduced to merely arranging the spot addresses in proper order.
Thus, context is recognized.
"Offsets" are the spot addresses relative to the icon address.
All offsets are defined according to an icon address of 0 (see Figure
17).
Image
Border
Spot
Addresses
Figure 17. Image addresses and offsets,
image edges.
Some addresses overlap

44
Pattern geometry affects the upper limits of measurable strain
because large strain displacements may change the order of the off¬
sets in the deformed image. Spreading spots further apart (in the
undeformed image) increases the size of a spot's subregion. A larger
subregion allows the spot more strain displacement before any change
in offset order occurs.
The details are best illustrated by example (see Figure 18). It
is assumed that the undeforraed image is analyzed first and, with one
exception, both images receive the same preliminary analysis.
i
Undeformed
Figure 18.
A simple example of motion (3u./3X .=0.05, u=v=-10 pixels,
0=120°, E. . =0.0525). The origin ^is in the upper left
corner of1¿ach image.

45
Successful mapping is initially assumed and later verified by four
final checks.
First, the image is segmented and the features are extracted.
Scanning is by column and the spot centroids are stored in the order
they were found. (See columns marked "Raw" in Tables 1 and 2.) The
icon is identified, and it's rotation is also computed.
Table 1. Unordered spot coordinates in the undeformed image.
Raw
Reference
From
Icon
X
Y
X
Y
X
Y
58.0101
57.9660
57.4541
57.8310
-70.5459
-70.1690
93.0079
58.0056
92.4517
57.9438
-35.5483
-70.0562
163.0518
57.9950
162.4955
58.0797
34.4955
-69.9203
197.9896
58.0041
197.4332
58.1617
69.4332
-69.8383
127.9832
57.9801
127.4271
57.9914
-0.5729
-70.0086
53 .0481
93-0576
57.4187
92.9227
-70.5813
-35.0773
93-0216
92.9791
92.3923
92.9173
-35.6077
-35.0827
197.9616
93-0132
197.3321
93.1757
69.3321
-34.8243
127.9396
92 .9724
127.3103
92.9836
-0.6897
-35.0164
163.0107
92.9915
162.3813
93-0760
34.3813
-34.9240
128.7025
1 27 .9873
128.0000
128.0000
0 .0000
0.0000
57.9599
128.0263
57.2575
127.8911
-70.7425
-0.1089
93.0284
128.0002
92.3260
127.9383
-35.6740
-0.0617
198.0660
128.0575
197.3632
128.2151
69.3632
0.2151
63.0010
127.9894
162.2984
1 28.0738
34.2984
0.0738
92.9964
163.0297
92.2207
162.9676
-35.7793
34.9676
127.9581
162.9798
127.1824
162.9909
-0.8176
34.9909
162.9653
163.0147
162.1896
163.0989
34.1896
35.0989
197.9768
162.9531
197.2011
163.1105
69.2011
34.1105
57.9856
162.9875
57.2101
162.8523
-70.7899
34.8523
58 .06 35
198.0315
57.2148
197.8964
-70.7852
69.8964
92.9969
193.0169
92.1481
197.9548
-35.8519
69.9548
128.0099
198.0273
127.1610
198.0384
-0.8390
70.0384
163.0027
198.0136
162.1538
198.0978
34.1538
70.0978
197.9936
197.9982
197.1446
198.1556
69.1446
70.1556
NOTE: Units are pixels. "Raw" coordinates are the spot centroids
in the image. "Reference" and "From Icon" values are computed by
equations (46) and (47), respectively. See Figure 18.

45
Table 2. Unordered spot coordinates in the deformed image.
Raw
Ref erence
From
Icon
X
Y
X
Y
X
Y
98.5336
33.4158
51.9634
192.1880
-76.0366
64.1880
131 .2724
50.2682
53.0816
155.3833
-74.9184
27.3833
78.6761
64.3727
88.4532
196.7857
-39.5468
68.7857
163.9343
67.1092
54.2228
118.6530
-73.7772
-9.3470
111.3736
81.1960
89.5630
160.0309
-38.4370
32.0309
196.6288
84.0388
55.4298
81 .8551
-72.5702
-46.1449
58.7113
95.2677
124.9335
201.5070
-3.0665
73.5070
144.0612
98.0816
90.7333
123.2581
-37.2667
-4.7419
229.4016
100.8559
56.5016
45.0350
-71 .4984
-82.9650
91.4833
112.1065
126.0252
164.6782
-1 .9748
36 .6782
176.7262
114.9435
91 .8919
86.5161
-36.1081
-41.4839
123.7393
1 29.6776
128.0000
128.0000
0.0000
0.0000
38.9059
126.2522
161.4258
206.0458
33.4258
78.0458
209.4617
131 .8162
93.0300
49 .7056
-34.9700
-78.2944
71.5500
143.1225
162.6010
169.3190
34.6010
41 .3190
156.8817
145.8710
1 28.3497
91 .1147
0.3497
-36.8853
18.9873
157.2419
197.9716
210.6846
69.9716
82.6846
104.3150
160.0107
163.7403
132.4753
35.7403
4.4753
189.5733
162.8025
129.5597
54.3186
1 .5597
-73.6814
51 .6854
174.0585
199.0750
173.9321
71 .0750
45.9321
136.9841
176.8253
164.8545
95.7498
36.8545
-32.2502
84 .3163
190.9335
200.2602
137.2151
72.2602
9.2151
169.6761
193.7303
166 .0404
58.9647
38 .0404
-69.0353
117.1079
207.7552
201 .3230
100.3761
73.3280
-27 .6239
149.7807
224.6135
202.4800
63.6284
74.4800
-64.3716
NOTE: Units are pixels. "Raw" coordinates are the spot centroids
in the image. "Reference" and "From Icon" values are computed by
equations (46) and (47), respectively. See Figure 18.
All spots are moved into a "reference position" by translation
and rotation (see Figure 19). Translation is accomplished by adding
X^ to the X and Y^ to the Y of each spot; X^ and Y^ are
X_ = X -x
T cp Icon
Y = Y -Y
T cp Icon
(45)

47
where X and Y are the coordinates of the icon in reference posi-
cp cp F
tion (typically the image center). Rotation for each spot is
computed from the x and y distance between it and the icon. The
computation is
x = x-xT
p Icon
Y = Y-Y
Icon
â–  V
2 2
X + Y
P P
Tan 1(^)
XRef = X+p[cos(u)z-9)-cos a>z]+XT
YRef = Y+p[sin((jjz-0)-sin wz]+YT
(45)
where XRg^ and YRe^> are the new coordinates of each spot in the
reference position. The results of translation and rotation are
shown in Tables 1 and 2 under the "Reference" column.
Position relative to the icon, i.e., distance from the icon, is
x = x - Y
Rel Ref Icon
Y = Y - Y
Rel Ref Icon
(47)

48
Figure 19- The deformed image and its reference position.
The distances from the icon are shown in Tables 1 and 2 under the
columns marked "From Icon." Although all spots have been located
relative to the icon, recognizing the pattern they form and arranging
them in proper order (by address) is not always easy--especially if
rotation is more than a few degrees.
Address computation is based on "learning" the undeformed
image. The spots nearest and farthest from the icon row and column
are located in the undeformed image and used to calculate the address
offset," Aoffset, by the following:

49
where
X = MAX(X)
max
X . = MIN(X)
mm
Y = MAX(Y)
max
Y . = MIN(Y)
min
N = NINT(tt^)
x X
min
N = NINT(y^)
y min
X -N (X . )
X x + max x mm
space min N +1
Y -N (Y . )
Y y + max y min
space min N +1
y
(48)
Nshift = NINT (f^_) ♦ 1
space
(49)
X Y
Aoffset â–  NINT space
space
(50)
Xmax = maximum X centroid
Xmin = minimum X centroid
Ymax = maximum Y centroid
Ymin = minimum Y centroid
= number of rows above or below icon

50
Ny = number of columns left or right of icon
Nr0Ws = number of rows in the image
X3pacs = approximate row spacing
YSpace = approximate column spacing
NShift = number of possible columns in image.
'Function NINT rounds numbers to the nearest integer value. Values
Xspace’ Yspace’ and Nshift are saved fn0111 the undeformed image and
used in the deformed.
Fortunately, arranging the spots in proper sequence is a simple
matter of bubble sorting [73] the offsets in ascending order. By
equation (50), offsets (see column "Offset" in Tables 3 and 4)
increase down columns and left to right. Column "Key" is the order
in which the spots were found. Note, the two-dimension pattern
context is now represented by a one-dimension array, X , which
corresponds to the ordered spot addresses.
Recognition of the two-dimension context is accomplished by
determining the row-column order of the spots. The left-most column
is assumed to begin with the lowest offset and continue with ascend¬
ing offset until the X coordinate decreases. Column 2 begins with
the next highest offset and continues until the X coordinate again
decreases. The process continues until the offset array is exhaust¬
ed. By incrementing row and column counters, the relative spot order
is determined which establishes context (see Tables 5-8).
Until now, both undeformed and deformed images were processed
the same (except for learning the undeformed image) and assumed to
have the same context. This assumption is verified by the following

51
Table 3-
Ordered
spot coordinat
es in the
undeformed image.
Reference
From
Icon
Offset
Key
X
Y
X
Y
-18
1
57.4541
57.8310
-70.5459
-70.1690
-17
2
92.4517
57.9438
-35.5483
-70.0562
-16
5
127.4271
57.9914
-0.5729
-70.0086
-15
3
162.4955
58.0797
34.4955
-69.9203
-14
4
197.4332
58.1617
69.4332
-69.8383
-10
6
57.4187
92.9227
-70.5813
-35.0773
-9
7
92.3923
92.9173
-35.6077
-35.0827
-8
9
127.3103
92.9836
-0.6897
-35.0164
-7
10
162.3813
93.0760
34.3813
-34.9240
-6
8
197.3321
93.1757
69.3321
-34.8243
-2
12
57.2575
127.8911
-70.7425
-0.1089
-1
13
92.3260
1 27.9383
-35.6740
-0.0617
0
11
128.0000
128.0000
0.0000
0.0000
1
15
162.2984
128.0738
34.2984
0.0738
2
14
197.3632
128.2151
69.3632
0.2151
6
20
57.2101
162.8523
-70.7899
34.8523
7
16
92.2207
162.9676
-35.7793
34.9676
8
17
127.1824
162.9909
-0.8176
34.9909
9
18
162.1896
163.0989
34.1896
35.0989
10
19
197.2011
163.1105
69.2011
35.1105
14
21
57.2148
197.8964
-70.7852
69.8964
15
22
92.1481
197.9548
-35.8519
69.9548
16
23
127.1610
198.0384
-0.8390
70.0384
17
24
162.1538
198.0978
34.1538
70.0978
18
25
197.1446
198.1556
69.1446
70.1556
NOTE: Units are pixels. Offsets are computed by equation (50).
Keys are the order in which the spots were found. Undeformed
spacing is 35 pixels in x and y. See Figure 18.

52
Table 4.
Ordered
spot coordinates in the
deformed image,
Reference
From
Icon
Offset
Key
X
Y
X
Y
-18
9
56.5016
45.0350
-71.4984
-82.9650
-17
14
93.0300
49.7056
-34.9700
-78.2944
-16
19
129.5597
54.3186
1.5597
-73.6814
-15
23
166.0404
58.9647
38.0404
-69.0353
-14
25
202.4800
63.6284
74.4800
-64.3716
-10
6
55.4298
81.8551
-72.5702
-46.1449
-9
11
91.8919
86.5161
-36.1081
-41.4839
-8
16
128.3497
91 .1147
0.3497
-36.8853
-7
21
164.8545
95.7498
36.8545
-32.2502
-6
24
201 .3280
100.3761
73-3280
-27.6239
-2
4
54.2228
118.6530
-73.7772
-9.3470
-1
8
90.7333
123.2581
-37.2667
-4.7419
0
12
128.0000
128.0000
0.0000
0.0000
1
18
163.7403
132.4753
35.7403
4.4753
2
22
200.2602
137.2151
72.2602
9.2151
6
2
53.0816
155.3833
-74.9184
27.3833
7
5
89.5630
160.0309
-38.4370
32.0309
8
10
126.0252
164.6782
-1 .9748
36.6782
9
15
162.6010
169.3190
34.6010
41.3190
10
20
199.0750
173.9321
71 .0750
45.9321
14
1
51.9534
192.1880
-76.0366
64.1880
15
3
88.4532
196.7857
-39.5468
68.7857
15
7
124.9335
201.5070
-3.0565
73.5070
17
13
161 .4258
206.0458
33.4258
78.0458
18
17
197.9716
210.6846
69.9716
82.6846
NOTE: Units are pixels. Offsets are computed by equation (50).
Keys are the order in which the spots were found. See Figure 18.

53
Table 5. Spot distance from origin in undeformed image.
COL
1
2
3
4
5
ROW
1
X:
58.0101
58.0481
57.9599
57.9856
58.0635
Y:
57.9660
93.0576
128.0263
162.9875
198.0315
COL
1
2
3
4
5
ROW
2
X:
93.0079
93-0216
93.0284
92.9964
92.9969
Y:
58.0056
92.9791
128.0002
163 .0297
198 .0169
COL
1
2
3
4
5
ROW
3
X:
127.9832
127.9396
128.7025
127.9581
128.0099
Y:
57.9801
92.9724
127.9873
162.9798
198.0273
COL
1
2
3
4
5
ROW
4
X:
163.0518
163.0107
163.0010
162.9653
163.0027
Y:
57.9950
92.9915
127.9894
163.0147
198.0136
COL
1
2
3
4
5
ROW
5
X:
197.9896
197.9616
198.0660
197.9768
197.9936
Y:
58.0041
93.0182
128.0575
162.9531
197.9982
NOTE: Units are pixels. The origin is the upper left corner of the
image. See Figure 18.

54
Table 6. Spot distance from origin in deformed image.
COL
1
2
3
4
5
ROW
1
X:
229.4016
196.6288
16-3.9343
131.2724
98.5336
Y:
100.8559
84.0388
67.1092
50.2682
33.4158
COL
1
2
3
4
5
ROW
2
X:
209.4617
176.7262
144.0612
1 1 1.3736
78.6761
Y:
131 .8162
114.9435
98.0816
81 .1960
64.3727
COL
1
2
3
4
5
ROW
3
X:
189.5733
156.8817
123.7393
91.4833
58.7113
Y:
162.8025
145.8710
129.6776
112.1065
95.2677
COL
1
2
3
4
5
ROW
4
X:
169.6761
136.9841
104.3150
71.5500
38.9059
Y:
193.7303
176.8253
160.0107
143.1225
126.2522
COL
1
2
3
4
5
ROW
5
X:
149.7307
117.1079
84.3163
51.6854
18.9873
Y:
224.6135
207.7552
190.9335
174.0585
157.2419
NOTE:
Units are
pixels. The
origin is
the upper
left corner
of the
image. See Figure 18.

55
Table 7. Spot distance from icon in undeformed image.
COL
1
2
3
4
5
ROW
1
X:
-70.5459
-70.5813
-70.7425
-70.7899
-70.7852
Y:
-70.1690
-35.0773
-0.1089
34.8523
69 .896 4
COL
1
2
3
4
5
ROW
2
X:
-35.5483
-35.6077
-35.6740
-35.7793
-35.8519
Y:
-70.0562
-35.0827
-0.0617
34.9676
69.9548
COL
1
2
3
4
5
ROW
3
X:
-0.5729
-0.6897
0.0000
-0.8176
-0.8390
Y:
-70.0086
-35.0164
0.0000
34.9909
70.0384
COL
1
2
3
4
5
ROW
4
X:
34.4955
34.3813
34.2984
34.1896
34.1538
Y:
-69.9203
-34.9240
0.0738
35.0989
70.0978
COL
1
2
3
4
5
ROW
5
X:
69.4332
69.3321
69.3632
69.2011
69.1446
Y:
-69.8383
-34.8243
0.2151
35.1105
70.1556
NOTE:
Units are
pixels. Spacing is 35
pixels in
x and y.
See
Figure 18.

56
Table 8. Spot distance from icon in deformed image.
COL
1
2
3
4
5
ROW 1
X:
-71.4984
-72.5702
-73.7772
-74.9184
-76.0366
Y:
-82.9650
-46.1449
-9.3470
27.3833
64.1880
COL
1
2
3
4
5
ROW 2
X:
-34.9700
-36.1081
-37.2667
-38.4370
-39.5468
Y:
-78.2944
-41 .4839
-4.7419
32.0309
68.7857
COL
1
2
3
4
5
ROW 3
X:
1.5597
0.3497
0.0000
-1.9748
-3.0665
Y:
-73*6814
-36.8853
0.0000
36.6782
73.5070
COL
1
2
3
4
5
ROW 4
X:
38.0404
36.8545
35.7403
34.6010
33.4258
Y:
-69.0353
-32.2502
4.4753
41 .3190
78.0458
COL
1
2
3
4
5
ROW 5
X:
74.4800
73.3280
72.2602
71.0750
69.9716
Y:
-64.3716
-27.6239
9.2151
45.9321
82.6845
NOTE:
Units are
pixels. See
Figure 18.

57
checks:
1. same number of rows in each image
2. same number of columns in each image
3. same number of spots in each image
4. rows times columns equal number of spots.
Of course, if any check fails, the images are "out of con¬
text." If strains are large enough, context similarities between
images cannot be recognized and an error message is printed. Assum¬
ing large, local rotation does not occur in the material, recognizing
similar context between the two patterns that does not exist is very
remote. Thus if no check fails, the approximation algorithm estab¬
lishes a one-to-one relationship, or "context mapping," between spots
in the two images.

CHAPTER III
IMAGE GENERATION AND ANALYSIS
III.1 Synthetic Images
Computer generated images, or "synthetic" images, were used as
the primary development tool and to study the influence of various
parameters. The options data file shown in Table 9 was used to gen¬
erate these 256X256 pixel images.
Motion is limited to affine transformations plus rigid-body
rotation about the image center. The transformations are
x =? c.X + c„Y + c„
1 2 3
y = c„X + cY + c,
450
(51)
There are three image output options available. The "BINARY"
option is the machine representation of the image. "DECIMAL" is the
machine version converted to a readable form (see, for example,
Figure 4b). "PLOT" displays a thresholded image, suitable for hard
copy, on a graphics terminal (see, for example, Figure 15).
"U123.DAT," "NIMG.DAT," "D123.DAT," and "DIMG.DAT" are user-defined
file names. The image generator automatically reads the input file
and generates the specified output thereby allowing the user a choice
of interactive or batch processing.
58

59
Table 9. Input file for synthetic image generator.
IMAGE GENERATOR INPUT DATA
SPOT GEOMETRY PATTERN SIZE
RADIUS
XSPACING
YSPACING
XCOLS
YROWS
4.00
35.00
35.00
5
5
GRAY LEVELS
BLACK WHITE
00000 00200
NOISE DATA
DISTRIBUTION TYPE: N
STANDARD DEVIATION: 007.000
NORMAL UNIFORM
IMAGE MOTION
MOTION TYPE: 2
1—POLYNOMIAL ONLY
2—POLYNOMIAL AND ROTATION
3—-ROTATION ONLY
POLYNOMIAL X
Y
TRANSLATION
x: +1.05000
+0.05000
-010.0000
y: +0.05000
+1.05000
-010.0000
ROTATION (DEGREES): +120
.000
IMAGE OUTPUT
FORMAT UNDEFORMED
Y/N DEFORMED Y/N
BINARY: U123.DAT Y
DECIMAL: NIMG.DAT N
PLOT: *SCREEN* N
D123.DAT Y
DIMG.DAT N
«SCREEN* N

60
III.2 Image Analysis
Similar to image generation, image analysis begins with an op¬
tions file which is shown in Table 10. If "USE THESE FILE NAMES?" is
answered no, the user is automatically prompted for the file name.
Since no other information is required by the program, the user is
free to choose interactive or batch processing.
Table 10. Image analysis options file.
******** IMAGE ANALYSIS INPUT/OUTPUT OPTIONS ********
INPUT SPECIFICATIONS
IMAGE FILE NAME:
USE THESE FILE NAMES?:
UNDEFORMED DEFORMED
+ +
U123 .DAT D123.DAT
N N
IMAGE OUTPUT UNDEFORMED DEFORMED
UNORDERED SPOT DATA: Y
DISTANCE FROM ORIGIN: Y
DISTANCE FROM ICON: Y
Y
Y
Y
DISPLACEMENT ANALYSIS OUTPUT
TOTAL DISPLACEMENT: Y
DISPLACEMENT GRADIENTS: Y
LAGRANGIAN STRAIN: Y
TAYLOR SERIES STRAIN: Y
MOTION ANALYSIS OUTPUT
DEFORMATION GRADIENT: Y
GREEN DEFORM. TENSOR: Y
RIGHT STRETCH TENSOR: Y
ROTATION TENSOR: Y
LAGRANGIAN STRAIN: Y

61
After segmentation, feature extraction, and context recognition,
automatic processing continues with motion analysis. The motion of
each spot is resolved and the displacement is computed by equation
(3) (see Table 11). Next, the deformation and displacement gradients
at each spot are computed by differentiating the functions used to
describe motion and displacement.
Table
11 .
Displacement of
each spot.
COL
1
2
3
4
5
ROW 1
X:
171.3916
138.5808
105.9744
73.2868
40.4701
Y:
42.8899
-9 .0188
-60.9171
-112.7193
-164.6158
COL
1
2
3
4
5
ROW 2
X:
116.4538
83.7046
51.0328
18.3772
-14.3208
Y:
73.8106
21 .9643
-29.9186
-81.8337
-133 .6442
COL
1
2
3
4
5
ROW 3
X:
61.5900
28.9421
-4.9632
-36.4747
-69.2986
Y:
104.8223
52.3985
1 .6903
-50.3733
-102 .7596
COL
1
2
3
4
5
ROW 4
X:
6.6243
-26.0267
-58.6861
-91.4154
-124.0968
Y:
135.7353
83.8338
32.0212
-19.8922
-71 .7614
COL
1
2
3
4
5
ROW 5
X:
-48.2088
-80.8538
-113.7497
-146.2914
-179.0063
Y:
166 .6094
11 4.7370
62.8761
11 .1054
-40.7563
NOTE: Units are pixels. Compare images in Figure 18. Displacements
are computed using the data in Tables 5 and 5 in equation (3).
The choice of a mapping model depends, of course, on the
application.
Intuition, a priori knowledge,
elasticity theory,

62
numerical efficiency, etc. are used to select the approximating func¬
tions. In this study, simple polynomials, linear or second order,
adequately approximate the motion and displacement in the synthetic
and real images. The models are linear
x =
G11X
+ c
21
V
y
°12X
+ c
22
V
u =
c X + c + c Y
13 23 33
v =
°14X + °24
+ V
(52)
and second order
x
C11X
C21X
'31
C41y2 +
c Y +
51
c,.XY
61
y
c
C22X
'32
+ C42y2 +
°52Y +
°62XY
u
°13x2 * °23X * °33 * W2 * °53Y * °63XI
V
'u*2 *
°24X + C34
C44y2 +
C54Y +
°64XY
(53)
These models are fitted to a small neighborhood of spots and
differentiated to obtain the displacement and deformation

63
gradients. Typically, the neighborhoods are 3X3, 5X5, or 7X7 and the
gradients are computed for the center spot. Exceptions are the spots
near the edge of the pattern and their gradients are computed like a
forward-backward difference.
For both models, the coefficients are determined by least
squares evaluation. Based on a modified form of the Conte and Deboor
[7-4] notation, the normal equations are
ÍÍ 2 =
(54)
where
Linear
(55)
X. Y„ X„Y
11
2 2
.. y;
.. Y.
•• xnyn
Second
Order

64
c =
'31
Linear
Second
order
(56)
f =
y1
{ V u V
N N N
(57)
and N is the number of spots in the neighborhood. Let
A
and
$f = Z
Equations (54) become
Ac = Z
(58)
which are solvable by Gauss-Jordan elimination.

65
Hornbeck [75] points out Gauss-Jordan elimination may be used to
solve equations (58) , but unfortunately the set is extremely
ill-conditioned. Taking his advice, double precision arithmetic is
used to cope with the large variation in the magnitudes of the co¬
efficients in any given row. In addition to his suggestions,

are scaled before solution and the c's are reconverted during
gradient computation.
For a variety of programming reasons, solution of equations (58)
is not performed directly by Gauss-Jordan elimination. Instead, the
inverse of A is computed using Hornbeck's [75] Gauss-Jordan matrix
inversion subroutine, with the addition of appropriate error traps,
and c is solved by
o = A"1Z (59)
How accurate are the values of c and how well do equations (52)
or (53) fit the data? Debugging experience indicates the number of
reliable digits (numerically speaking, not experimentally) runs in
the teens. In this study the solutions are known, but if they were
unknown, goodness-of-fit and the contribution of each term could be
easily tested by the methods suggested by Miller and Freund [63].
The example of Figure 18 is continued with the calculation of
c. Assume the linear model of equation (52) and a 3X3 neighborhood
centered at spot (3.3) - A substitution of data from Table 5 into
equation (55) produces

66
93.0216
1 .0000
92 .9791
93-0284
1.0000
128.0002
92.9964
1 .0000
163 .0297
127.9396
1.0000
92.9724
128.7025
1 .0000
127.9873
127.9581
1.0000
162.9798
163.0107
1 .0000
92.9915
163.0010
1.0000
127.9894
162.9653
1 .0000
163.0147
Likewise, data from Tables 6 and 11 in equation (57) lead to
176.7262
114.9435
83.7046
21 .9643
144.0612
98.0816
51.0328
-29.9186
1 1 1 .3736
81 .1960
18.3772
-81 .8337
156.8817
145.8710
28.9421
52.8985
f =
123.7393
1 29.6776
-4.9632
1 .6903
91.4833
112.1065
-36.4747
-50.8733
136.9841
176.8253
-26.0267
83.8338
104.3150
160.0107
-58 .6861
32.0212
71 .5500
143.1225
-91 .4154
-19.8922
The solution for 0 is
-0.56849 0.88427
-1.56849
0.88427
c =
316.52130 77.50840
316.52130
77.50840
-0.93425 -0.48175
-0.93425
-1.48165
(61)
(62)
The motion functions (i.e., the "mathematical" maps) in equa¬
tions (52) become
x = -0.56849X + 316.52130 - 0.93425Y
(63)
y = 0.88427X + 77.50840 - 0.48175Y

57
and the displacement maps in equations (52) become
u = -1.56849X + 316.52130 - 0.93425Y
v = 0.88427X + 77.5084 - 1 .48175Y
(64)
The deformation and displacement gradients, computed by equa¬
tions (2) and (4), respectively, become
F
-0.56849
0.88427
-0.93425
-0.48175
J =
-1 .56849
0.88427
-0.93425
-1.48175
Strain, computed by equations (11) or (12), is
E =
0.05255
0.05256
0.05256
0.05246
(65)
(66)
(67)
By the Taylor series method of equation (13). strain is
E
0.05124
0.04763
0.04763
0.05115
(68)

68
Strain computed according to small displacement theory is, in
this case, quite inaccurate. By equations (14),
-1.56849 -0.02499
-0.02499 -1.48175
(69)
Rigid-body rotation is computed as follows. By equation (8),
the Green deformation tensor is
C =
1 .10509
0.10511
0.10511
1.10491
The eigenvalues of C are
C =
-P
1 .21009
0.00000
0.00000
0.99990
1 /2
and C is
-P
1.10004 0.00000
0.00000 0.99995
Since U = C1/2, then
-P ~P
U =
1 .05000
0.05005
0.05005
1.05000
(70)
(71 )
(72)
(73)

69
Rotation tensor R, computed by equation (17), is
R
-0.50016
0.86596
-0.86596
-0.50011
The remainder of the output for the example in Figure
(74)
18 is
shown in Tables 12 through 19.

70
Table 12. Displacement gradients.
COL
1
2
3
4
5
ROW
1
3u/3X
3v/3X
3u/3Y
3v/3Y'
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1 .56849
0.88427
-0.93425
-1 .48175
-1 .56849
0.88427
-0.93425
-1 .48175
COL
1
2
3
4
5
ROW
2
3u/3X
3v/3X
3u/3Y
3v/3Y
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1 .56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
COL
1
2
3
4
5
ROW
3
3u/3X
3v/3X
3u/3Y
3v/3Y
-1 .56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .43175
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
COL
1
2
3
4
5
ROW
4
3u/3X
3v/3X
3u/3Y
3v/3Y
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1 .56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1 .56849
0.88427
-0.93425
-1 .48175
COL
1
2
3
4
5
ROW
5
3u/3X
3v/3X
3u/3Y
3v/3Y
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93^25
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
-1.56849
0.88427
-0.93425
-1 .48175
NOTE: Correct values are 3u/9X = -1.56830, 3v/9X = 0.88433,
3u/3Y = -O.93433, 3v/3Y = -1.48170.

71
Table 13. Lagrangian strain from displacement gradients.
COL
1
2
3
4
5
ROW 1
^11
52554.
52554.
52554.
52554.
52554.
E12
E22
52557.
52557.
52557.
52557.
52557.
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 2
E1 1
52554.
52554.
52554.
52554.
52554.
E12
E22
52557.
52557.
52557.
52557.
52557.
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 3
E11
52554.
52554.
52554.
52554.
52554.
E ?
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 4
E11
52554.
52554.
52554.
52554.
52554.
E '
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 5
E11
52554.
52554.
52554.
52554.
52554.
E1?
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
NOTE:
52,500.
Units
are micro inches
-per-inch.
Correct
values are
EU '

72
Table 14. Strain from Taylor series.
COL
1
2
3
4
5
ROW 1
^11
51241.
51241.
51241 .
51241.
51241.
E
47634.
47634.
47634.
47634.
47634.
1 C\J
â–  c\]
IÚ
51151 .
51151.
51151.
51151.
51151.
COL
1
2
3
4
5
ROW 2
E11
51241 .
51241 .
51241 .
51241 .
51241 .
E12
47634.
47634.
47634.
47634.
47634.
E22
51151 .
51151 .
51151 .
51 151 .
51151 .
COL
1
2
3
4
5
ROW 3
E11
51241.
51241.
51241.
51241.
51241.
E1 2
47634.
47634.
47634.
47634.
47634.
E22
51151.
51151 .
51151.
51151 .
51151.
COL
1
2
3
4
5
ROW 4
E11
51241 .
51241 .
51241 .
51241 .
51241 .
E1 2
47634.
47634.
47634.
47534.
47634.
E22
51151 .
51151 .
51 151 .
51151 .
51151 .
COL
1
2
3
4
5
ROW 5
E11
51241 .
51241.
51241 .
51241 .
51241.
E12
47634.
47634.
47634.
47634.
47634.
E22
51151 .
51151.
51 151 .
51151.
51151.
NOTE: Units are micro inches
= 51 ,190, E12 = E21 = 47,582.
-per-inch.
Correct
values are
E11 = E22

73
Table 15. Deformation gradient.
COL
1
2
3
4
5
ROW 1
F11
F1 2
F21
F22
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
COL
1
2
3
4
5
ROW 2
F11
p12
F21
F22
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
COL
1
2
3
4
5
ROW 3
F11
F1 2
F21
F22
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
COL
1
2
3
4
5
ROW 4
M2
F21
F22
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
COL
1
2
3
4
5
ROW 5
F11
F1 2
f21
22
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
-0.56849
-0.93425
0.88427
-0.48175
NOTE: Correct values are
0.88433, F22 = -0.48170.
Fn = -0.56830, F12 =
-0.93433, F21
=

74
Table 16. Green deformation tensor.
COL
1
2
3
4
5
ROW
1
C11
c12
C22
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
COL
1
2
3
4
5
ROW
2
C1 1
C1 2
C22
1 .1051 1
0.10511
1.10492
1 .1051 1
0.10511
1 .10492
1 .10511
0.10511
1.10492
1 .10511
0.10511
1 .10492
1.10511
0.10511
1 .10492
COL
1
2
3
4
5
ROW
3
C11
C1 2
C22
1.10511
0.10511
1.10492
1.10511
0.1051 1
1.10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
COL
1
2
3
4
5
ROW
4
C11
C1 2
C22
1 .1051 1
0.10511
1.10492
1 .10511
0.10511
1.10492
1.10511
0.10511
1.10492
1 .10511
0.1051 1
1 .10492
1.10511
0.10511
1.10492
COL
1
2
3
4
5
ROW
5
C11
C1 2
C22
1.10511
0.10511
1.10492
1.1051 1
0.1051 1
1 .10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
1.10511
0.10511
1.10492
NOTE: Correct values are = C22 = 1.10500, C12 = 0.10500.

75
Table 17. Right stretch tensor.
COL
1
2
3
4
5
ROW
1
U11
u12
u22
1.05005
0.05005
1 .04996
1.05005
0.05005
1 .04996
1.05005
0.05005
1.04996
1.05005
0.05005
1.04996
1.05005
0.05005
1 .04996
COL
1
2
3
4
5
ROW
2
U11
U1 2
U22
1 .05005
0.05005
1 .04996
1 .05005
0.05005
1.04996
1.05005
0.05005
1 .04996
1 .05005
0.05005
1.04996
1 .05005
0.05005
1 .04996
COL
1
2
3
4
5
ROW
3
U11
U12
U22
1.05005
0.05005
1.04996
1.05005
0.05005
1 .04996
1.05005
0.05005
1.04996
1.05005
0.05005
1.04996
1 .05005
0.05005
1.04996
COL
1
2
3
4
5
ROW
4
U11
U12
U22
1.05005
0.05005
1.04996
1 .05005
0.05005
1.04996
1.05005
0.05005
1.04996
1 .05005
0.05005
1 .04996
1 .05005
0.05005
1 .04996
COL
1
2
3
4
5
ROW
5
U11
U1 2
U22
1.05005
0.05005
1 .04996
1.05005
0.05005
1.04996
1.05005
0.05005
1.04996
1 .05005
0.05005
1 .04996
1.05005
0.05005
1 .04996
NOTE: Correct values are U-, ^ = U22 = 1.05000, U-tp = 0.05000.

76
Table 18. Rotation tensor.
COL
1
2
3
4
5
ROW 1
R11
R1 2
R21
R22
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
COL
1
2
3
4
5
ROW 2
R11
R12
R21
R22
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
~0.86596
0.86596
-0.5001 1
-0.50011
-0.86596
0.86596
-0.5001 1
COL
1
2
3
4
5
ROW 3
R11
R1 2
R21
R22
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.5001 1
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
COL
1
2
3
4
5
ROW 4
R11
R1 2
R21
R22
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0 .86596
0.86596
-0.5001 1
COL
1
2
3
4
5
ROW 5
R11
R1 2
R21
R22
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.5001 1
-0.50011
-0.86596
0.86596
-0.50011
-0.50011
-0.86596
0.86596
-0.50011
NOTE: Correct values are
0.86603, R22 = ~°•50000.
R11 = -0.50000, R12 =
-0.86603, R21
=

77
Table 19. Lagrangian strain from deformation gradient.
COL
1
2
3
4
5
ROW 1
E11
52554.
52554.
52554.
52554.
52554.
E, p
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 2
E11
52554.
52554.
52554.
52554.
52554.
E1 2
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 3
E11
52554.
52554.
52554.
52554.
52554.
E1?
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 4
E11
52554.
52554.
52554.
52554.
52554.
E1 p
52557.
52557.
52557.
52557.
52557.
E22
52459.
52459.
52459.
52459.
52459.
COL
1
2
3
4
5
ROW 5
E11
52554.
52554.
52554.
52554.
52554.
E.p
52557.
52557.
52557.
52557.
52557.
E
¿2
52459.
52459.
52459.
52459.
52459.
NOTE:
52,500
Units
are micro inches
-per-inch.
Correct
values are
Eij "

CHAPTER IV
EXPERIMENTS
IV. 1 Test Specimen
The prismatic beam shown in Figure 20 was used for all experi¬
ments. A clear plastic, CR-39, was chosen for its relatively low
modulus of elasticity and capacity for large, elastic strain.
The 11X11 test pattern on the beam was generated by the synthe¬
tic image generator and applied by contact printing. First, a hard
copy of the pattern was photographed using Kodalith film. After
coating the beam with a buffer of polyurethane and naphtha, a Liquid
Light emulsion was applied, contact printed, and chemically fixed
(see Figure 21).
IV.2 Test Equipment
Three experimental set-ups were configured; two were for strain
tests and the third for rigid-body motion.
The strain set-up was a typical photoelastic load frame with
special fixtures (see Figure 22). The fixtures were for four-point
bending (see Figure 23) and cantilever beam (see Figure 24) experi¬
ments.
The equipment shown in Figure 25 was used for the rigid-body
motion tests. A photocopy of a protractor was taped to the back of
the lens holder and rotation was measured relative to the tip of the
78

79
— t—
Figure 20. Beam fabricated from CR-39.
Figure 21
Spot pattern on beam

80
Figure 22. Experimental set-up of load frame.
Figure 23. Set-up for four-point bending.

81
Figure 24. Set-up for cantilever beam.
Figure 25. Rigid-body motion set-up.

82
clamped pointer. Measurements were accurate to about 1 degree. A
dial indicator measured translation in the horizontal direction with
an accuracy of ±0.0005 inch.
Since image processing equipment was unavailable at the
University of Florida, the equipment for the rigid-body experiments
was transported to USC where the digital images were recorded.
IV.3 Experimental Procedure
Unfortunately, the strain tests were not done because the load
frame was too large to transport to USC. The second order model in
equation (53) was for these tests.
The equipment shown in Figure 25 and an image processing system
similar to the one shown in Figure 3 (based on the IBM PC) was used
for the rigid-body experiments. The CCD camera was a Sony model
XCM-38 with a zoom lens, and the digitizer board was an eight-bit
Datacube "Frame Grabber" with 385 rows and 374 columns of pixels.
The set-up included several routine checks. First, alignment
between the rail, table, and camera were checked to minimize experi¬
mental error. Then, the zoom lens was adjusted to enlarge the
pattern image as much as possible. Resolution in the horizontal
direction was measured at 358 pixels per inch.
Lighting was adjusted to provide the usual illumination for
tests in the USC lab, and the transparent specimen caused some
lighting problems. Contrast was impaired by the light transmitted
through the beam.

83
After set-up, images were digitized and stored on floppy disk.
All images were transferred to magnetic tape via USC's VAX-11/780
computer for transport to the University of Florida.

CHAPTER V
ANALYSIS OF RESULTS
The analysis of results has been divided into two sections:
numerical and experimental. The numerical section summarizes synthe¬
tic image work and the experimental section summarizes the analysis
of the images digitized at USC.
V.1 Numerical
The analysis of synthetic images included rigid-body motion,
homogeneous strain, combined motion, and the influence of noise, gray
levels, and spot radius. To ensure a good random sample, the results
from a variety of spot spacings were averaged. The pattern was 5X5
with the icon being spot (3,3). Except as noted, the gray level
difference was 200, and the zero-mean noise was Gaussian with a
standard deviation of 7.0 gray levels. The absolute values of noisy
pixels with values less than zero were used instead of the negative
gray levels.
In the following tables, the average absolute error is defined
as follows: average absolute error is the average of the absolute
values of the errors. For example, if and the computed
values are E^=100y, E^2=~’150y, and E22=50p, the average absolute
error is 100y.
84

85
Standard deviation, also in the tables, is the standard devia¬
tion of the computed values. For the strain example above, the
standard deviation is 132y.
For translations of various amounts in the x and y directions,
the average absolute error was 0.008 pixels with a standard deviation
of 0.011 pixels. Translations near whole pixels had slightly less
error. (Equation (38) suggested a spatial sensitivity.)
Rigid-body rotation was about the image center which was approx¬
imately two-thirds of a pixel below the pattern center. Rotations
ranged from 0 to ±180 degrees, and the results, computed using the
elements of the rotation tensor, were very consistent with an average
absolute error of 0.0039 degrees and a standard deviation of 0.0008
degrees.
Several strain cases were investigated and summarized in Table
20. Normal strains were ^ =E22 E12=E21=^; shear strains were
E11=E22=0 w^h E1 2=e21 ’ and combined strains were E-|-j =E22=E1 2=E21 ‘
See equation (12) for strain definitions.
Noise had a significant effect on results. As suggested by
equation (38), increasing SNR decreased translation and strain error
(see Table 21). Framing failures began occurring below SNR=100.
Gray level was another significant variable because of the SNR
and the digitization of illumination intensity. For constant noise,
increasing gray level difference increased the SNR and thereby in¬
creased accuracy. With a noise standard deviation of 7.00 gray
levels, spots were frequently unidentifiable below a gray level

36
Table 20. Summary of strain analysis for synthetic images.
Strain
Case
Displacement
Gradient
9V3xj
Strain
Correct
Ave. Abs.
Error
Standard
Deviation
Normal
0.000010
10
17
10
0.000100
100
17
15
0.001000
1001
22
27
0.010000
10050
40
32
0.100000
105000
40
49
Shear
0.000010
10
24
0
0.000100
100
29
0
0.001000
1000
25
0
0.010000
10000
33
0
0.090000a
90000
34
0
Combined
0.000010
10
3
14
0.000100
100
7
28
0.001000
1000
23
26
a Limited by pattern geometry.
NOTE: Rigid-body motion was zero. Strain units are micro inches-
per-inch. SNR=816.

87
Table 21. The effect of SNR on translation and strain calculations.
Translation
Strain
Ave. Abs.
Standard
Ave. Abs.
Standard
SNR
Error
Deviation
Error
Deviation
100
0.0049
0.0362
39
58
200
0.0029
0.0253
19
42
400
0.0021
0.0176
20
35
600
0.0016
0.0143
14
32
800
0.0012
0.0123
10
26
1000
0.0012
0.0109
8
21
2500
0.0008
0.0068
5
8
5000
0.0006
0.0043
3
6
10000
0.0004
0.0034
2
5
00
0.0001
0.0001
0
0
NOTE:
Translations were
10.5 pixels
in x and y. No
rotation. All
strains
were Eij. = 100y.
Table 22. The effect of
gray level
difference on translation and
strain calculations.
Translation
Strain
Ave. Abs.
Standard
Ave. Abs.
Standard
H
Error
Deviation
Error
Deviation
32
0.0257
0.0270
##
**
64
0.0043
0.0349
66
62
128
0.0026
0.0172
20
45
256
0.0009
0.0089
3
16
512
0.0007
0.0043
5
8
1024
0.0003
0.0022
4
5
2048
0.0002
0.0011
2
2
4096
0.0001
0.0006
1
2
NOTE:
Translations were
10.5 pixels
in x and y. No
rotation. All
strains
were E^-ioOp.

88
difference of 32. Below a difference of 64, a strain of Eij=100y was
undetectable (see Table 22).
An increase in gray level difference reduced the error caused by
digitization. Without noise, a difference of 16 gray levels produced
an average absolute strain error of 12y (for Eij=100p) with a stand¬
ard deviation 0.03ij and, above a difference of 64, error was nil.
However, SNR effects greatly overshadowed digitization error.
The effect of spot radius decreased rapidly with increased
radius (see Table 23). Equation (38) suggested this for small E,,
and, in fact, almost no change occurred for radii above 10 pixels.
Table 23. The effect of spot radius on translation and strain
calculations.
Translation
Strain
Spot
Ave. Abs.
Standard
Ave. Abs.
Standard
Radius
Error
Deviation
Error
Deviation
2
0.0030
0.0233
29
20
4
0.0018
0.0272
10
32
6
0.0013
0.0085
7
15
8
0.0016
0.0069
8
13
10
0.0020
0.0052
8
19
12
0.0021
0.0059
5
11
NOTE:
strains
Translations were 10
were Eij=100y. SNR=
.5 pixels
816.
in x and y. No
rotation. All
How much VAX 11/750 computer time did the analysis of a 256X256
pixel image require? For a 3X3 spot pattern, CPU time was about 4.9

89
seconds with no printout written to disk and about 6.5 seconds with a
full printout (see Tables 1-9 and 12-19). An 11X11 pattern required
about 14.9 seconds for no printout and about 28.3 seconds for a full
printout. Due to overall system load, CPU times varied as much as 15
percent.
V.2 Experimental
Images digitized at USC were for rigid-body experiments. These
included rotations of 1.0, 10.0, and 63.5 degrees; horizontal (y
direction) translations of 1.00, 10.00, and 10.50 pixels; and a com¬
bined motion of 30.0 degrees rotation plus 10.00 pixels translation
in the horizontal direction. The images digitized during the experi¬
ments were summarized in Table 24.
Table 24. Images used for rigid-body motion experiments.
Image
Gray
Black
Level
White
an
H
SNR
Undeformed
U005
57
103
3.72
46
153
U006
88
151
5.16
63
149
U007
50
95
2.49
45
327
Deformed
D005
91
150
2.98
59
392
D006
96
158
3.56
62
303
D007
96
1 58
3.18
62
380
D008
82
1 47
4.57
65
202
D009
81
145
4.42
64
210
D01 0
86
151
4.26
65
233
D011
48
90
2.80
42
225
NOTE: Gray levels,
a , and H
n
were estimated
by the analysis
program.

90
In spite of relatively low SNR, the images were still usable.
The SNR for the images were near the minimum usable level suggested
by synthetic image analysis. In fact, undeformed images U005 and
U006 required a simple, ad hoc adjustment to one of their mapping
parameters. Otherwise, analysis proceeded as expected.
Rotations, accomplished using the lens holder previously de¬
scribed, were about a point near the image center. As shown in Table
25, the average results, except for the third test, were within the 1
degree accuracy of the experiment.
The average and the average absolute strain error was computed
using all the strains (i.e., E-|-j , E12> and E22 at a11 1 sP°ts)
computed by the analysis program. The actual strain for any
Table 25. Results of rigid-body rotation experiments.
Image
Rotation
Strain
Error
Actual
Ave.
Error
Std. Dev.
Ave.
Ave. Abs.
D005
1 .00
0.81
-0.19
0.03
-119
336
D006
10.00
9.23
-0.77
0.02
360
529
D007
63.50
64.83
1.33
0.03
191
3737
D011
30.00a
30.49
0.49
0.02
1661
2298
a 30.00 degrees plus 10.00 pixels horizontal translation.
NOTE: The undeformed image was U005 for the first three experiments
and U007 for the last. All rotations were in degrees. Strain units
are micro inches-per-inch. Measurement accuracy was about 1 degree.

91
rigid-body motion was, of course, 0.0. Although the noise was
Gaussian, the average strain error was not the expected value of
0.0. The reason for this was imaging system distortion.
Average results obtained during translation experiments are
summarized in Table 26. The first test was within the ±0.18 pixel
resolution of the set-up and the other two tests were not. Evidently
there was a slight "tilt" between the axes of the lens holder and the
camera.
There was good, linear consistency between the three translation
tests. The quotient of the second translation divided by 10.0 was
1.041 and the third by 10.5 was 1.047. This consistency suggested
some systematic error in the experimental measurement probably caused
the increased error in the latter tests.
An apparent "distortion" existed in all images, and the effect
of this was manifested as high strain error. This was especially
noticeable in the average strain error which had an expected value of
0.0. Since the tests were not preceeded by extensive calibration,
the exact source cannot be positively identified. No doubt the dis¬
tortion was induced by the image processing system.
Other sources of error were low SNR, low gray level difference,
and relatively coarse resolution of the experiment set-up. Although
a common solution for minimizing noise effects is averaging multiple
images, only one copy of each image was used in these experiments. A
small increase in gray level difference was possible through the use
of additional lighting (or by using an opaque, reflective

Table 26. Results of rigid-body translation experiments.
Image
Translation
Strain Error
Horizontal
Vertical
Actual
Ave.
Error
Std. Dev.
Actual
Ave.
Error
Std. Dev.
Ave.
Ave. Abs.
D008
1 .000
1 .045
0.045
0.043
0.000
0.008
0.008
0.040
218
368
D009
10.000
10.405
0.405
0.058
0.000
0.016
0.016
0.058
-13
196
D010
10.500
10.994
0.494
0.053
0.000
0.041
0.041
0.029
-107
347
NOTE:
The undeformed
image
was U00Ó and
all translations
are in
pixels. Strain
units
are micro
inches-per-inch. Measurement accuracy was ±0.18 pixels.
r\j>

93
specimen). Improved resolution was not possible due to the limita¬
tions of available equipment.
The analysis program made efficient use of computer resources.
Typical CPU time for a real image with no printout written to disk
was about 25 seconds and about 38 CPU seconds for full printout.
Although not implemented in this study, it was possible to
further decrease CPU time. For example, a simple strain definition
such as change in gage length divided by gage length and estimating
rigid-body rotation by icon rotation would have saved about 12 CPU
seconds. Additional savings of about 6 CPU seconds per deformed
image (for images with the same undeformed image) were possible if
the undeformed image were analyzed only once.

CHAPTER VI
CONCLUSIONS AND RECOMMENDATIONS
VI.1 Conclusions
The numerical and experimental results successfully demonstrated
the potential of pattern mapping in motion analysis. With this
method, it was possible to obtain accurate measurements of rotation,
translation, and strain for a variety of plane motions (see Tables
20-26). Automation minimized user time and required only a modest
amount of computer resources. Whole image analysis was also
thoroughly demonstrated.
Segmentation algorithms performed well in all images tested.
The estimation of histogram parameters was highly efficient and pro¬
vided adequate results without user interaction. Except for very low
SNR, automatic framing was easily accomplished.
Mapping algorithms, designed to produce unique mapping, worked
well for a large variety of motions. Although low SNR prevented
automatic mapping in some images, no erroneous maps were produced.
In two of the experimental images, low SNR necessitated a manual
reset of one of the parameters. Averaging multiple images or
filtering may have helped the SNR problem.
The most important variables were SNR and gray level difference
with higher values yielding better accuracy. Numerical and
94

95
experimental results suggested a practical minimum SNR between 100
and 200. Although a lack of data prevents a direct comparison of
numerical and experimental results, they agreed "directionally" for
rotation and translation.
In general, spot radius was not an important variable. Practi¬
cal experience and numerical results indicated radii should be at
least 4 to 6 pixels.
Pattern mapping has the potential for significant data compres¬
sion. Other image processing applications require archiving entire
images. If the pattern is well suited to the application, the entire
displacement field may be adequately reconstructed from the spot
information.
VI.2 Recommendations
Although the experimental data included in this study is encour¬
aging, pattern mapping needs extensive experimental verification with
a wide variety of plane motion applications. Only by using a rigor¬
ously calibrated system and analyzing many images, can the full
potential of this method be known.
A logical next step is the consideration of image processing and
pattern recognition tools such as filtering, local averaging, other
dimensions of feature space, etc. These may significantly improve
performance in low SNR situations or lead to the discovery of new
parameters and more general mapping theories. The implementation of
the full syntactic recognition theory should also be investigated.

96
An extension of this method to three-dimensional motion and
mapping of multiple images (mosaics) is also suggested. Three-
dimensional images that are not restricted by camera view have much
potential for practical application.
As a final note, it is suggested that pattern mapping be inte¬
grated with other experimental methods and numerical analysis
tools. This includes advanced graphics for data manipulation and
presentation.

APPENDIX
BORDER FOLLOWING ALGORITHM

ISTARTl
í>
(KX-1 , YXlf) .AND. (I(X-1 ,Y-1)>K)
+ NO
(I(X-1,Y+1)It)
+ NO
(KX.Y+1 )It)
4- NO
(KX + 1 ,Y+1 )It)
4- NO
(I(X+1 ,Y)It)
4- NO
(KX+1 , Y-1 Xlf ) .AND.(I(X4-1 ,Y)>It )
4- NO
KX.Y-1 ) YES
YES
YES
YES
YES
YES
YES
X=X-1 >1
x=x-1,Y=Y+1
>
Y=Y+1
X=X+1,Y=Y+1
X=X + 1
X=X+1,Y=Y-1
Y= Y-1
4- NO
I(X-1,Y-1) 4- NO
KX-1 ,Y) YES
YES
X=X-1, Y=Y-1
X=X-1
+ NO
FRAME ERROR: STOP
YES
(X=X0).AND.(Y=Y0)
END
4- NO
YES
I(X,Y)>0
I(X,Y)=-I(X,Y)
+ NO
I(X,Y)=I(X,Y)+BIAS
YES
>@
98

REFERENCES
1. Dally, J. W., and Riley, W. F., Experimental Stress Analysis,
McGraw-Hill, New York (1973).
2. Durelli, A. J., Phillips, E. A., and Tsao, C. H., Introduction
to the Theoretical and Experimental Analysis of Stress and
Strain, McGraw-Hill, New York (1958).
3. Richards, C. E., Engineering Materials Science, Wadsworth, San
Francisco (1961).
A. Vest, C. M., Holographic Interferometry, John Wiley, New York
(1978).
5. Taylor, C. E., Recent Trends in Optical Methods, Developments in
Mech., 13, 327-330, Proc. Nineteenth Mid-Western Mech. Conf.,
Columbus, OH (September 1985).
6. Taylor, C. E., Applications of Coherent Optics to Experimental
Mechanics, Proc. Int. Sym. on Photoelasticity, Tokyo (May 1986).
7. Burger, C. P., Automated Moire and Photoelastic Analysis with
Digital Image Processing, Proc. 1984 SEM Fall Conf., Milwaukee,
WI (November 4-7, 1984).
9.Chen, T. Y., Application of Digital Image Processing and
Computer Graphics Techniques in Photomechanics, Ph.D. Disserta¬
tion, Univ. of Florida, Gainesville (1985).
9. Chu, T. C., Digital Correlation Method in Experimental
Mechanics, Ph.D. Dissertation, Univ. of S. Carolina, Columbia
(1982).
10. McNeill, S. R., and Sutton, M. A., A Development for the Use of
Measured Displacements in Boundary Element Modelling, Eng.
Anal., 2, 124-127 (1985).
11. Robinson, D. W., Automatic Fringe Analysis with a Computer
Image-Processing System, Appl. Opt., 22, 2169-2175 (1983).
12. Seguchi, Y., Tomita, Y., and Watanabe, M., Computer-Aided
Fringe-Pattern Analyzer—A Case of Photoelastic Fringe, Exp.
Mech., 19, 362-370 (1979).
99

100
13. Voloshin, A. S., and Burger, C. P., Half-fringe Photoelasti¬
city: A New Approach to Whole-Field Stress Analysis, Exp.
Mech., 23, 304-313 (1983).
14. Weathers, J. M., Foster, W. A., Swinson, W. F., and Turner, J.
L., Integration of Laser-Speckle and Finite-Element Techniques
of Stress Analysis, Exp. Mech., 25, 60-65 (1985).
15. Gonzales, R. C., and Wintz, P., Digital Image Processing,
Addison-Wesley, Reading, MA (1977).
16. Mendenhall, F. T., Computer Aided Analysis of Photoelastic
Images, Ph.D. Dissertation, Univ. of Illinois, Urbana (1981).
17. PDP 11/70 Processor Handbook, Digital Equipment Corp., Maynard,
MA (1976).
18. Rosenfeld, A., Digital Picture Processing, vols. 1 and 2,
Academic Press, New York (1982).
19. Merrill, P. S., Photodot Investigation of Plastic Strain
Patterns in Flat Sheet with a Hole, Exp. Mech., 1, 73-80 (1961).
20. Chu, T. C., Ranson, W. F., Sutton, M. A., and Peters, W. H.,
Applications of Digital-Image-Correlation Techniques to Experi¬
mental Mechanics, Exp. Mech., 25, 232-244 (1985).
21. Dalloul, S. A., Image Correlation in Deformable Bodies Including
Non-linear Geometries, Ph.D. Dissertation, Univ. of S. Carolina,
Columbia (1984).
22. Ranson, W. F., Digital Image Processing, Proc. 1984 SEM Fall
Conf., Milwaukee, WI (November 4-7, 1984).
23. Sutton, M. A., Wong, A. K., and Chao, Y. J., Enhanced Displace¬
ment Measurement for Double-Aperture Specklegrams, Exp. Mech.,
23, 348-353 (1983).
24. Sutton, M. A., Wolters, W. J., Peters, W. H., Ranson, W. F., and
McNeil, S. R., Determination of Displacements Using an Improved
Digital Correlation Method, Image Vision Comput., 1, 133-139
(1983).
25. Sigler, D., and Haworth, W. L., Strain Measurement by Optical
Correlation, J. Nondest. Eval., 2, 125-132 (1982).
26. Box, W. A., The Effect of Plastic Strains on Stress Concentra¬
tions, Proc. SESA, 8, 99-110 (1951).
27. Durelli, A. J., and Riley, W. F., Developments in the Grid
Method of Experimental Stress Analysis, Proc. SESA, 9, 91-100
(1957).

101
28. Miller, J. A., Improved Photogrid Techniques for Detection of
Strain Over Short Gage Lengths, Proc. SESA, 10, 29_34 (1952).
29. Obata, M., Shimada, H., and Kawasaki, A., Fine-Grid Method for
Large Strain Analysis Near a Notch Tip, Exp. Mech., 23, 146-151
(1983).
30. Obata, M., and Shimada, H., Embedded Fine-Grid Method Applied to
Three-Dimensional Stress-Intensity-Factor Analysis, Exp. Mech.,
25, 8-11 (1985).
31. Parks, V. J., The Grid Method, Exp. Mech., 9, 27N-33N (1969).
32. Parks, V. J., Strain Measurement Using Grids, Opt. Eng., 21,
633-639 (1982).
33- Sevenhuijesen, P. J., We Need Simplicity and Speed, Presented at
BSSM/RAS Joint Conf., NLR, Amsterdam (1982).
34. Sevenhuijesen, P. J., Two Simple Methods for Deformation Demon¬
stration Measurement, Strain, 17, 20-24 (1981).
35. Ballard, D. A., and Brown, C. M., Computer Vision, Prentice-
Hall, Englewood Cliffs, NJ (1982).
36. McVey, E. S., and Lee, J. W., Some Accuracy and Resolution
Aspects of Computer Vision Distance Measurements, IEEE Trans.
PAMI, PAMI-4, 646-649 (1982).
37. Muri, A. Z., On Monocular Perception of 3“D Moving Objects, IEEE
Trans. PAMI, PAMI-2, 582-585 (1980).
38. Roach, J. W., and Aggarwal, J. K., Determining the Movement of
Objects from a Sequence of Images, IEEE Trans. PAMI, PAMI-2,
554-562 (1980).
39. Huang, T. S., and Tsai, R. Y., 3“D Motion Estimation from Image-
Space Shifts, IEEE Proc. ICASSP, 3, 1136-1139 (1981).
40. Tsai, R. Y., and Huang, T. S., Estimating Three-Dimensional
Motion Parameters of a Rigid Planar Patch, IEEE Trans. ASSP,
ASSP-29, 1 147-1152 (1981 ).
41. Tsai, R. Y., and Huang, T. S., Estimating Three-Dimensional
Motion Parameters of a Rigid Planar Patch, II: Singular Value
Decomposition, IEEE Trans. ASSP, ASSP-30, 525-534 (1982).
42. Tsai, R. Y., and Huang, T. S., Estimating Three-Dimensional
Motion Parameters of a Rigid Planar Patch, III: Finite Point
Correspondences and the Three-View Problem, IEEE Trans. ASSP,
ASSP-32, 213-219 (1984).

102
43. Tsai, R. Y., and Huang, T. S., Uniqueness and Estimation of
Three-Dimensional Motion Parameters of a Rigid Objects, IEEE
Trans. PAMI, PAMI-6, 13*26 (1984).
44. Eringen, A. C., Mechanics of Continua, Wiley, New York (1967).
45. Leigh, D. C., Nonlinear Continuum Mechanics, McGraw-Hill, New
York (1968).
46. Malvern, L. E., Introduction to the Mecnanics of a Continuous
Medium, Prentice-Hall, Englewood Cliffs, NJ (1969).
47. Trusdell, C., and Toupin, R., The Classical Field Theories,
Encyclopedia of Physics, vol. III/1, 226-309, Springer-Verlag,
Berlin (i960).
48. Trusdell, C., and Noll, W., The Non-linear Field Theories of
Mechanics, Encyclopedia of Physics, vol. III/3, 48-56, Springer-
Verlag, Berlin (1965).
49. Green, W. B., Digital Image Processing: A Systems Approach, Van
Nostrand, New York (1983).
50. Duff, M. J. B., ed., Languages and Architectures for Image
Processing, Academic Press, New York (1981).
51. Kittler, J., and Duff, M. J. B., eds., Image Processing System
Architecture, Research Studies Press (Wiley), New York (1985).
52. Rosenfeld, A., Picture Languages, Academic Press, New York
(1979).
53. Green, W. B., Jepsen, P. L., Krezner, J. E., Ruiz, R. M.,
Schwartz, A. A., and. Seidman, J. B., Removal of Instrument
Signature from Mariner 9 Television Images of Mars, Appl. Opt.,
14, 105-114 (1975).
54. Fu, K. S., and Mui, J. K., A Survey on Image Segmentation, Pat.
Rec., 13, 3-16 (1981 ).
55. Leboucher, G., and Lowitz, G. F., What a Histogram Can Really
Tell the Classifier, Pat. Rec., 10, 351-357 (1978).
56. Dunn, S. M., Harward, D., and Davis, L. S., Local Estimation of
the Uniform Error Threshold, IEEE Trans. PAMI, PAMI-5, 742-747
(1984).
57. Kittler, J., Threshold Selection Based on a Simple Image
Statistics, Comput. Vision Graphics Image Process., 30, 125-147
(1985).
58. Nakagawa, Y., and Rosenfeld, A., Some Experiments on Variable
Thresholding, Pat. Rec., 11, 191-204 (1979).

103
59.
60.
Ó1 .
62.
63.
64.
65.
66.
67.
68.
69.
Otsu, N., A Threshold Selection Method From Gray-Level Histo¬
grams, IEEE Trans. SMC, SMC-9, 62-66 (1979).
Weszka, J. S., Nagel, R. N., and Rosenfeld, A., A Threshold
Selection Technique, IEEE Trans. Comput., C-23, 1322-1326
(1974).
Weszka, J. S., and Rosenfeld, A., Threshold Evaluation Techni¬
ques, IEEE Trans. SMC, SMC-8, 622-629 (1978).
Weszka, J. S., and Rosenfeld, A., Histogram Modification for
Threshold Selection, IEEE Trans. SMC, SMC-9, 38-52 (1979).
Miller, I., and Freund, J. E., Probability and Statistics for
Engineers, Second Ed., Prentice-Hall, Englewood Cliffs, NJ
(1977).
Fukunaga, K., Introduction to Statistical Pattern Recognition,
Academic Press, New York (1972).
Gonzalez, R. C., and Thomason, M. G., Syntactic Pattern Recogni¬
tion, Addison-Wesley, Reading, MA (1978).
Pratt, W. K., Digital Image Processing, Wiley, New York (1978).
Duda, R. 0., and Hart, P. E., Pattern Classification and Scene
Analysis, Wiley, New York (1973).
Tou, J. T., and Gonzalez, R. C., Pattern Recognition Principles,
Addison-Wesley, Reading, MA (1974).
Prewitt, J. M. S., and Mendelsohn, M. L., The Analysis of Cell
Images, Ann. N. Y. Acad. Sci., 128, 1035-1053 (1966).
70. Fu, K. S., Syntactic Pattern Recognition and Applications,
Prentice-Hall, Englewood Cliffs, NJ (1982).
71. Volker, C., E'nrig, H., and Rozenburg, G., eds., Lecture Notes in
Computer Science, vol. 73, Springer-Verlag, Berlin (1979).
72. Shi, Q. Y., and Fu, K. S., Parsing and Translation of (Attri¬
buted) Expansive Graph Languages for Scene Analysis, IEEE Trans.
PAMI, PAMI-5, 472-485 (1983).
73. Brooks, F. P., Jr., and Iverson, K. E., Automatic Data
Processing, Wiley, New York (1969).
74. Conte, S. D., and Deboor, C., Elementary Numerical Analysis,
Third Ed., McGraw-Hill, New York (1980).
75. Hornbeck, R. W., Numerical Methods, Prentice-Hall, Englewood
Cliffs, NJ (1975).

BIOGRAPHICAL SKETCH
R. Wallace Fail was born in Mobile, Alabama, on August 3,
1954. He grew up in southern Alabama and graduated from Fort Dale
Academy in 1972.
He graduated from The University of Alabama in 1977 with a
B.S.M.E. and in 1979 with a M.S.M.E. While at the University, he was
a member of Tau Beta Pi, Pi Tau Sigma, ASME, and ASHRAE. In addi¬
tion, Wallace was a co-op student with Alabama Power Company at
Greene County Steam Plant near Demopolis, Alabama.
In 1979, Wallace joined Chevron, U.S.A., at the Pascagoula
Refinery, in Pascagoula, Mississippi. He enjoyed several assignments
with Chevron before he returned, in 1983, to school at the University
of Florida to pursue a Ph.D. in engineering mechanics.
Wallace is married to Robyn, an educator and Ph.D. student in
special education at the University of Florida.
104

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
(1a1 ^CU^¿'q
c"! E. Taylor, Chairman
Professor of Engineering Sciences
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
4-- 0 ck~Q -£u-x-
S. S. Ballard
Distinguished Service Professor of
Physics
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
U. H. Kurzweg
Professor of Engineering Sciences

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
E. K. Walsh
Professor of Engineering Sciences
This dissertation was submitted to the Graduate Faculty of the
College of Engineering and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08553 9939



xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EUONPDKRW_YBY7Q0 INGEST_TIME 2012-02-07T15:14:26Z PACKAGE AA00003790_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

3$77(51 0$33,1* ,1 3/$1( 027,21 $1$/<6,6 %< 5 :$//$&( )$,/ $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

&RS\ULJKW E\ 5 :DOODFH )DLO

PAGE 3

7R WKH *ORU\ RI *RG 'LJLWL]HG E\ WKH ,QWHUQHW $UFKLYH LQ ZLWK IXQGLQJ IURP 8QLYHUVLW\ RI )ORULGD *HRUJH $ 6PDWKHUV /LEUDULHV ZLWK VXSSRUW IURP /\UDVLV DQG WKH 6ORDQ )RXQGDWLRQ KWWSZZZDUFKLYHRUJGHWDLOVSDWWHUQPDSSLQJLQ22IDLO

PAGE 4

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

PAGE 5

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,, 6\QWKHWLF ,PDJHV ,PDJH $QDO\VLV Y

PAGE 6

,9 (;3(5,0(176 ,9 7HVW 6SHFLPHQ ,9 7HVW (TXLSPHQW ,9 ([SHULPHQWDO 3URFHGXUH 9 $1$/<6,6 2) 5(68/76 9 1XPHULFDO 9 ([SHULPHQWDO 9, &21&/86,216 $1' 5(&200(1'$7,216 9, &RQFOXVLRQV 9, 5HFRPPHQGDWLRQV $33(1',; %25'(5 )2//2:,1* $/*25,7+0 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ YL

PAGE 7

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

PAGE 8

7KH HIIHFW RI JUD\ OHYHO GLIIHUHQFH RQ WUDQVODWLRQ DQG VWUDLQ FDOFXODWLRQV 7KH HIIHFW RI VSRW UDGLXV RQ WUDQVODWLRQ DQG VWUDLQ FDOFXODWLRQV ,PDJHV XVHG IRU ULJLGERG\ PRWLRQ H[SHULPHQWV 5HVXOWV RI ULJLGERG\ URWDWLRQ H[SHULPHQWV 5HVXOWV RI ULJLGERG\ WUDQVODWLRQ H[SHULPHQWV YLLL

PAGE 9

/,67 2) ),*85(6 )LJXUH 3DJH 7KH PRWLRQ RI ERG\ WKURXJK VSDFH 7KH PRWLRQ RI WKH QHLJKERUKRRG RI SDUWLFOH S $ W\SLFDO 3&EDVHG LPDJH SURFHVVLQJ V\VWHP $ VSRW DQG LWV GLJLWDO LPDJH $ SL[HO DQG LWV HLJKWQHLJKERUV 7KH ELPRGDO KLVWRJUDP RI ILJXUH %ODFN DQG ZKLWH UHJLRQV LQ D ELPRGDO KLVWRJUDP )UDPH RI VSRW LQ ILJXUH ,PDJH DUWLIDFWV 2QHGLPHQVLRQ PRGHO 7KH GLVFUHWL]DWLRQ RI DQG NG L ,FRQ URWDWLRQ PHDVXUHG IURP WKH YHUWLFDO ,FRQ URWDWLRQ LV GHWHUPLQHG WR ZLWKLQ D FRQVWDQW E\ HTXDWLRQ f 3DWWHUQV WKDW DUH HDVLO\ UHFRJQL]HG DQG DQDO\]HG 3DWWHUQV RI )LJXUH ZLWK ,FRQV $ SDWWHUQ DQG LWV FRQWH[WVHQVLWLYH ODQJXDJH ,PDJH DGGUHVVHV DQG RIIVHWV $ VLPSOH H[DPSOH RI PRWLRQ 7KH GHIRUPHG LPDJH DQG LWV UHIHUHQFH SRVLWLRQ %HDP IDEULFDWHG IURP &5 ,;

PAGE 10

6SRW SDWWHUQ RQ EHDP ([SHULPHQWDO VHWXS RI ORDG IUDPH 6HWXS IRU IRXUSRLQW EHQGLQJ 6HWXS IRU FDQWLOHYHU EHDP 5LJLGERG\ PRWLRQ VHWXS [

PAGE 11

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n GRWV 7KHVH SDWWHUQV ZHUH DSSOLHG WR WKH VSHFLPHQ E\ FRQWDFW SULQWn LQJ $Q LFRQ ZDV LQFOXGHG LQ WKH SDWWHUQ WKHUHE\ DFFRPPRGDWLQJ ULJLGERG\ URWDWLRQV XS WR RQH UHYROXWLRQ 0RWLRQ IXQFWLRQV ZHUH VHOHFWHG E\ WKH XVHU DQG HDVLO\ FKDQJHG WR VXLW WKH SDUWLFXODU DSSOLFDWLRQ )XQFWLRQ FRHIILFLHQWV ZHUH GHWHUn PLQHG E\ OHDVW VTXDUHV [L

PAGE 12

$FFXUDF\ GHSHQGHG SULPDULO\ RQ WKH VLJQDOWRQRLVH UDWLR DQG WKH JUD\ OHYHO GLIIHUHQFH EHWZHHQ EODFN DQG ZKLWH $Q LQFUHDVH LQ WKHVH YDULDEOHV LPSURYHG DFFXUDF\ 8VHU DQG FRPSXWHU WLPH UHTXLUHPHQWV ZHUH TXLWH PRGHVW ,PDJHV ZHUH SURFHVVHG ZLWK OLWWOH RU QR XVHU LQWHUDFWLRQ DQG RXWSXW VSHFLn ILFDWLRQV ZHUH DXWRPDWLFDOO\ UHDG IURP D GLVN ILOH 'HSHQGLQJ RQ RXWSXW VSHFLILFDWLRQV FRPSXWHU WLPH 9$; f IRU DQ ; SDWWHUQ LQ D ; LPDJH ZDV WR &38 VHFRQGV [LL

PAGE 13

&+$37(5 ,1752'8&7,21 %DFNJURXQG 7KH SULPDU\ SXUSRVH LQ H[SHULPHQWDO VROLG PHFKDQLFV LV WR GHWHUn PLQH VWUHVV DQG PRWLRQ VWUDLQ ULJLGERG\ WUDQVODWLRQ DQG URWDWLRQf IURP PHDVXUHPHQWV RI SK\VLFDO TXDQWLWLHV )RU H[DPSOH VWUHVV PD\ EH FRPSXWHG IURP PHDVXUHPHQWV RI ORDG DQG FURVVVHFWLRQDO DUHD DQG VWUDLQ LV PHDVXUHG E\ FKDQJHV LQ FKDUDFWHULVWLF OHQJWKV 0DQ\ PHWKRGV PHFKDQLFDO HOHFWULFDO RSWLFDO HWFf DUH DYDLODEOH WR DQDO\]H VWUHVV DQG PRWLRQ >@ 2SWLFDO PHWKRGV SOD\ DQ LPSRUWDQW UROH LQ H[SHULPHQWDO PHFKDQn LFV 7KH\ KDYH D QXPEHU RI DGYDQWDJHV ZKLFK LQFOXGH >@ 1RQFRQWDFWLQJ PHDVXUHPHQWV GR QRW LQWHUIHUH ZLWK PDWHULDO UHVSRQVH )XOOILHOG DQDO\VLV LV XVXDOO\ SRVVLEOH +LJKO\ DFFXUDWH UHVXOWV DUH XVXDOO\ SRVVLEOH 5HVSRQVH WLPHV DUH DV IDVW DV OLJKW WKHUHE\ DFFRPPRGDWLQJ G\QDPLF LQYHVWLJDWLRQV $FWXDO VWUXFWXUHV XQGHU OLYH ORDGV FDQ IUHTXHQWO\ EH DQDn O\]HG (IIHFWV RI DFWXDO ERXQGDU\ FRQGLWLRQV FDQ EH VWXGLHG LQ GHWDLO

PAGE 14

$OWKRXJK RSWLFDO PHWKRGV RIIHU PDQ\ DGYDQWDJHV WKH\ GR KDYH VHYHUDO XQGHVLUDEOH IHDWXUHV )RU H[DPSOH VRPH PHWKRGV DUH VHQVLn WLYH WR VPDOO H[WUDQHRXV YLEUDWLRQV DQG UHTXLUH H[SHQVLYH GHOLFDWH HTXLSPHQW 3HUKDSV WKH PRVW XQGHVLUDEOH IHDWXUH LV WKH IDFW WKDW PRVW RI WKHVH PHWKRGV GR QRW GLUHFWO\ \LHOG WKHLU ZHDOWK RI LQIRUPDn WLRQ DERXW VWUHVV RU PRWLRQ LQ D UHDGLO\ XVDEOH IRUP 7KH H[WUDFWLRQ RI WKLV LQIRUPDWLRQ LV WLPHFRQVXPLQJ DQG UHn TXLUHV VSHFLDO VNLOOV $Q H[SHUW PD\ VSHQG KRXUV UHGXFLQJ GDWD DQG DOO WRR RIWHQ WKH DPRXQW RI GDWD DQDO\]HG LV OLPLWHG VLPSO\ EHFDXVH DXWRPDWHG DQDO\VLV LV QRW DYDLODEOH )XUWKHUPRUH UHVXOWV FDQ EH DIIHFWHG E\ WKH LQYHVWLJDWRUnV SDUWLFXODU WHFKQLTXHV 'LJLWDO FRPSXWHUV DUH RSHQLQJ GRRUV WR DXWRPDWLRQ LQFUHDVHG DFFXUDF\ RI H[LVWLQJ PHWKRGV DQG SUHYLRXVO\ LQWUDFWDEOH RU XQNQRZQ WHFKQLTXHV 6RPH RI WKH ODWHVW LQQRYDWLRQV HPSOR\ GLJLWDO LPDJH SURFHVVLQJ DQG SDWWHUQ UHFRJQLWLRQ ,335f WHFKQRORJLHV ZKLFK XWLOL]H GLJLWDO FRPSXWHUV DQG YLGHR LPDJLQJ HTXLSPHQW IRU GLUHFW GDWD DFTXLn VLWLRQ 7KH ILHOGV RI ,335 DUH UDSLGO\ H[SDQGLQJ DQG DUH ZHOO HVWDEOLVKHG LQ PDQ\ DUHDV ,Q H[SHULPHQWDO PHFKDQLFV WKHLU SRWHQn WLDO IRU HIILFLHQWO\ JDWKHULQJ YDVW TXDQWLWLHV RI GDWD FRQVLVWHQWO\ DQG DFFXUDWHO\ LV EHLQJ TXLFNO\ GHYHORSHG 7KHVH WHFKQRORJLHV DUH IUHHLQJ H[SHULPHQWDOLVWV IURP KRXUV RI SDLQVWDNLQJ ODERU DQG SURYLGn LQJ WKHP ZLWK D QHZ DYHQXH WR H[LVWLQJ QXPHULFDO PHWKRGV DQG DQDO\WLn FDO WRROV >@

PAGE 15

, 3XUSRVH RI 3UHVHQW :RUN 7KH ,335 WHFKQRORJLHV KDYH WKHLU SULFH )RU H[DPSOH ,335 SURJUDPV WKDW XVH IDVW )RXULHU WUDQVIRUPV RU FRUUHODWLRQ WHPSODWH PDWFKLQJf DUH IUHTXHQWO\ FRQVLGHUHG FRPSXWDWLRQDOO\ H[SHQVLYH EHFDXVH WKH\ HQWDLO ODUJH QXPEHUV RI UHODWLYHO\ VORZf IORDWLQJSRLQW FDOFXn ODWLRQV ,QWHUDFWLYH SURJUDPV SURYLGH JUHDW IOH[LELOLW\ EXW WKH\ QHHG WKH XVHUnV FRQVWDQW DWWHQWLRQ ,WHUDWLYH DQG EDFNWUDFNLQJ V\VWHPDWLFDOO\ UHWUDFLQJ SUHYLRXV VWHSVf PHWKRGV FDQ UHTXLUH VXEn VWDQWLDO DPRXQWV RI FRPSXWHU PHPRU\ IRU LQWHUPHGLDWH GDWD VWRUDJH >@ 6LJQLILFDQWO\ UHGXFLQJ WKHVH WLPH DQG PHPRU\ UHTXLUHPHQWV LV FHUWDLQO\ D GHVLUDEOH JRDO 3HUKDSV WKLV LV SRVVLEOH E\ UHWKLQNLQJ WKH SUREOHP DQG GHYHORSLQJ D KLJKO\ DXWRPDWHG PHWKRG ZKLFK FDSLWDOn L]HV RQ WKH VWUHQJWKV RI WKH GLJLWDO FRPSXWHU 7KXV WKH SXUSRVH RI WKLV ZRUN LV WR GHYHORS VXFK D PHWKRG SDWWHUQ PDSSLQJ WR REWDLQ GDWD YLD GLJLWDO LPDJLQJ HTXLSPHQW DQG VXEVHTXHQWO\ GHWHUPLQH SODQH PRWLRQ 6FRSH RI 3UHVHQW :RUN $OWKRXJK LW ZDV QRW DSSDUHQW LQ WKH EHJLQQLQJ WKH FRQFHSWV GHYHORSHG LQ WKLV LQYHVWLJDWLRQ DUH UHDGLO\ DSSOLHG WR WKH SKRWRGRW JULGf PHWKRG >@ 7KH VFRSH LQFOXGHV WKLV SKRWRGRW DGDSWDWLRQ $FWXDOO\ WKLV LQYHVWLJDWLRQ HQFRPSDVVHV D YDULHW\ RI VFLHQWLILF GLVFLSOLQHV 3HUKDSV WKH EHVW DSSURDFK LV WR FRQVLGHU WKH DWWULEXWHV RI HDFK DQG ZRUN WRZDUG D JRDO WKDW LV REWDLQDEOH LQ D UHDVRQDEOH DPRXQW RI WLPH 6SHFLILF REMHFWLYHV LQFOXGHG LQ WKH VFRSH DUH WR

PAGE 16

JXDUDQWHH XQLTXH PDSSLQJ SURYLGH VXESL[HO UHJLVWUDWLRQ RI SL[HO SURYLGH D ZLGH UDQJH RI PRWLRQ PHDVXUHPHQW SURYLGH ZKROHLPDJH DQDO\VLV UHGXFH FRPSXWHU DQG XVHU WLPH DV PXFK DV SRVVLEOH SURYLGH KLJKO\ DXWRPDWHG DQDO\VLV 3L[HOV DUH GLVFUHWH HOHPHQWV LQ D GLJLWDO LPDJH DQG ZKROHLPDJH LPSOLHV UHVXOWV DUH FRPSXWHG IRU HYHU\ SKRWRGRWf 6SHFLDO HPSKDVLV LV RQ DFFXUDF\ DQG LQFUHDVHG FRPSXWDWLRQ HIILFLHQF\ 6HYHUDO ODERUDWRU\ H[SHULPHQWV XVLQJ D &5 SODVWLFf SULVPDWLF EHDP ZHUH LQFOXGHG LQ WKH VWXG\ 7KH H[SHULPHQWV ZHUH WUDQVODWLRQ URWDWLRQ JHQHUDO ULJLGERG\ PRWLRQ IRXUSRLQW EHQGLQJ DQG FDQWLOHYHU EHDP $OO H[SHULPHQWV ZHUH VWDWLF FDVHV RI PRWLRQ LQ D SODQH QRUPDO WR WKH RSWLFDO D[LV $ UHFWDQJXODU ;f SDWWHUQ RI SKRWRGRWV ZDV FRQWDFW SULQWHG RQ WKH ODWHUDO VLGH RI WKH EHDP )RU FRQYHQLHQFH PRWLRQ LV GHILQHG RQO\ LQ WHUPV RI WKH /DJUDQJLDQ IRUPXODWLRQ (YDOXDWLRQ RI PRWLRQ IXQFWLRQV DUH OLPLWHG WR D OLQHDU DQG D VHFRQG RUGHU DSSUR[LPDWLRQ 7KH OHDVW VTXDUHV PHWKRG RI FXUYH ILWWLQJ LV XVHG H[FOXVLYHO\ DQG QRUPDO HTXDWLRQV DUH VROYHG E\ PDWUL[ LQYHUVLRQ XVLQJ *DXVV-RUGDQ HOLPLQDWLRQ 6\QWKHWLF LPDJHV DUH XVHG WR GHEXJ DQDO\VLV SURJUDPV DQG LQYHVn WLJDWH WKH LQIOXHQFH RI YDULRXV IDFWRUV 0RWLRQ LQ WKH LPDJHV LQn FOXGH DIILQH WUDQVIRUPDWLRQV DQG ULJLGERG\ URWDWLRQ DERXW WKH LPDJH FHQWHU ,QWHUSRODWLRQ RI EHWZHHQSL[HO LQWHQVLWLHV LV ELOLQHDU DQG V\VWHPLQGXFHG GLVWRUWLRQV DUH DVVXPHG HOLPLQDWHG GXULQJ SUHSURFHVVn LQJ 1RLVH KDV D ]HUR PHDQ DQG VHOHFWDEOH *DXVVLDQ RU XQLIRUP

PAGE 17

GLVWULEXWLRQV ,PDJH VL]H LV ; SL[HOV ZKLFK DUH XQVLJQHG ELWV RU OHVV 7KH FDPHUD SRVLWLRQ LV DVVXPHG IL[HG 6HJPHQWDWLRQ DQG IHDWXUH H[WUDFWLRQ GLYLGLQJ WKH LPDJH LQWR VXEUHJLRQV DQG REWDLQLQJ XVHIXO LQIRUPDWLRQf LQFOXGH QHLWKHU SUHILOn WHULQJ QRU KLVWRJUDPPLQJ JUD\ OHYHOVf DQG DUH EDVHG RQ WKH UDQGRP VHOHFWLRQ RI D IHZ SL[HOV 3URFHVVLQJ LV DXWRPDWLF DQG GLUHFW EHFDXVH NQRZQ SDWWHUQV LQ WKH LPDJH DUH DOZD\V FRPSOHWHO\ YLVLEOH DQG UHFRJQL]DEOH )HDWXUH VSDFH LQFOXGHV JUD\ OHYHOV DUHDV FHQWURLGV DQG VHFRQG PRPHQWV 7KH SDWWHUQ ODQJXDJH FRQVLVWV RI RQO\ RQH VHQn WHQFH DQG LW LV FRQWH[WVHQVLWLYH 6XUYH\ RI 3UHYLRXV :RUN 6ROLG 0HFKDQLFV (DUO\ DSSOLFDWLRQV RI LPDJH SURFHVVLQJ LQ H[SHULPHQWDO PHFKDQLFV LQYROYHG WKH DQDO\VHV RI IULQJHV 7KH IULQJHV ZHUH IURP SKRWRHODVWLn FLW\ PRLUH RU KRORJUDSKLF LQWHUIHURPHWU\ %XUJHU >@ DQG &KHQ >@ GHVFULEH VHYHUDO RI WKHVH DSSOLFDWLRQV LQ GHWDLO $QRWKHU LPDJH SURFHVVLQJ WHFKQLTXH DSSOLHG WR PHFKDQLFV LV FRUn UHODWLRQ 5HVHDUFKHUV >@ GHWHUPLQH SODQH GLVSODFHPHQWV XLI L f DQG WKHLU JUDGLHQWV 4XA2;M LM f E\ FRPSDULQJ GLJLWDOO\ UHFRUGHG LPDJHV RI WKH VDPH UDQGRP SDWWHUQ )RU H[DPSOH D SDWWHUQ RI EODFN VSHFNOHV HJ PLQXWH VSHFNV RI EODFN SDLQWf RQ D ZKLWH ILHOG LV UHFRUGHG DV LQWHJHU JUD\ OHYHOV DW GLVFUHWH SRLQWV SL[HOVf 7ZR LPDJHV DUH XVHG LQ WKH DQDO\VLV RQH EHIRUH DQG RQH DIWHU PRWLRQ %\ FRPSDULQJ VSDWLDOO\ UHJLVWHULQJf VPDOO VXELPDJHV IURP HDFK GLVSODFHPHQWV DQG WKHLU JUDGLHQWV DUH GHWHUPLQHG DW

PAGE 18

VHOHFWHG SRLQWV LQ WKH RULJLQDOf SDWWHUQ 6LJOHU DQG +DZRUWK >@ FRPELQH KRORJUDSK\ ZLWK FRUUHODWLRQ WR GHWHUPLQH PRWLRQ 7KXV WKH FRUUHODWLRQ PHWKRG KDV WKH DGYDQWDJHV RI XVLQJ FRKHUHQW RU FRQYHQWLRQDO OLJKW VRXUFHV DQG HDVLO\ DSSOLHG SDWWHUQV 7KH JULG PHWKRG LV RQH RI WKH ROGHVW DQG VLPSOHVW PHWKRGV XVHG LQ WKH DQDO\VLV RI VWUDLQ 2QH VLPSO\ DSSOLHV UHIHUHQFH PDUNV OLQHV VSRWV VFUDWFKHV HWFf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n JDWRUV GHILQH D PDUNnV SRVLWLRQ E\ LWV OHIW HGJH VRPH E\ LWV ULJKW HGJH DQG VRPH E\ D YDULHW\ RI GHILQLWLRQV $OO WKHVH GHILQLWLRQV DUH VXEMHFW WR HUURU FDXVHG E\ URWDWLRQ WKH QRQOLQHDU FKDUDFWHULVn WLFV RI WKH UHFRUGLQJ PHGLXP LI SKRWRJUDSKLFDOO\ UHFRUGHGf JULG DSSOLFDWLRQ WHFKQLTXHV OLJKWLQJ RSHUDWRU IDWLJXH HWF >@ 3DUNV >@ VXJJHVWHG FRPSXWHU WHFKQRORJ\ PLJKW LPSURYH WKH DFFXUDF\ RI WKH PHWKRG DQG ZRXOG FHUWDLQO\ UHGXFH WKH ODERU

PAGE 19

LQYROYHG 3UHYLRXV HIIRUWV WR IXOO\ FRPSXWHUL]H WKH JULG PHWKRG KDYH QRW EHHQ UHSRUWHG LQ WKH OLWHUDWXUH 6HYHQKXLMHVHQ >A@ UHSRUWHG VRPH IHDVLELOLW\ VWXGLHV ZLWK D 5HWLFRQ SKRWRGLRGH DUUD\ FDPHUD 8VLQJ DQ HLJKWELW GLJLWL]HU DQG GLJLWL]LQJ WKH VLJQDO IURP HDFK GLRGH KH REWDLQHG D UHVROXWLRQ RI D IHZ KXQGUHG PLFUR LQFQHVSHULQFK ,PDJHV ZHUH UHFRUGHG SKRWRJUDSKLn FDOO\ ZLWKRXW WKH DLG RI DQ LPDJH SURFHVVLQJ V\VWHP ,PDJH 3URFHVVLQJ3DWWHUQ 5HFRJQLWLRQ 6HYHUDO UHVHDUFKHUV DFWLYH LQ ,335 KDYH EHHQ LQWHUHVWHG LQ WZR DQG WKUHHGLPHQVLRQDO PRWLRQ DQDO\VLV 0RVW HIIRUWV KDYH FRQFHQWUDWn HG RQ WKH PRWLRQ RI ULJLG ERGLHV FORXGV WKH KXPDQ ERG\ RSWLFDO IORZ HWF 0RWLRQ SDUDPHWHUV ZHUH REWDLQHG E\ D YDULHW\ RI LWHUDWLYH SURFHVVHV >a@ +XDQJ DQG 7VDL >@ SURSRVHG D GLUHFW PHWKRG RI PDSSLQJ WKH WKUHHGLPHQVLRQDO PRWLRQ RI D SODQH XVLQJ WZR WLPHVHTXHQWLDO LPDJHV 0DSSLQJ SDUDPHWHUV ZHUH GHWHUPLQHG XQLTXHO\ E\ VROYLQJ D VHW RI OLQHDU HTXDWLRQV 8QIRUWXQDWHO\ PRWLRQ SDUDOOHO WR WKH RSWLFDO D[LV FRXOG QRW EH GHWHUPLQHG DQG VPDOO DSSUR[LPDWLRQV ZHUH PDGH IRU VWUDLQ DQG URWDWLRQ 0DWKHPDWLFDO SURRIV JXDUDQWHHLQJ XQLTXH PDSSLQJ EHWZHHQ LPDJHV ZHUH LQFOXGHG

PAGE 20

&+$37(5 ,, 7+(25(7,&$/ )281'$7,216 ,, 6ROLG 0HFKDQLFV 7KH GHVFULSWLRQ RI PRWLRQ XVHG KHUH LV OLPLWHG WR WKH /DJUDQJLDQ GHVFULSWLRQ DQG LV FRQVLVWHQW ZLWK WKH FODVVLFDO ILHOG WKHRULHV RI PRGHUQ FRQWLQXXP PHFKDQLFV >@ 7KH /DJUDQJLDQ GHVFULSWLRQ LV VXIILFLHQW WR GHWHUPLQH WKH GLUHFWLRQV RI WKH SULQFLSDO D[HV RI VWUDLQ DQG WKH PDJQLWXGHV RI WKH SULQFLSDO VWUHWFKHV ,W LV WKHUHn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n QDWHV DUH XVXDOO\ PHDVXUHG ZLWK UHVSHFW WR WKH VDPH FRRUGLQDWH D[LV 7KH PRWLRQ RI D SRLQW LV GHVFULEHG E\ WKH UHODWLRQ RI WKH FRRUGLQDWHV EHWZHHQ WKH WZR VWDWHV 0DWKHPDWLFDOO\ PRWLRQ LV H[n SUHVVHG E\ D RQHWRRQH PDSSLQJ [ V\PEROL]HG DV

PAGE 21

[ [;Wf RU [ L ;[O[[Wf f % )LJXUH 7KH PRWLRQ RI ERG\ J WKURXJK VSDFH 7KH GHIRUPDWLRQ JUDGLHQW RI WKH PRWLRQ [ LV GHILQHG DV WKH GHULYDWLYH RU JUDGLHQW RI [ DW ; DQG LV GHQRWHG E\ ) [ D[ RU ) LM f ZKHUH ) LV WKH IXQGDPHQWDO TXDQWLW\ IRU WKH DQDO\VLV RI ORFDO SURSHUn WLHV RI GHIRUPDWLRQ DQG WKH SK\VLFDO FRPSRQHQWV DUH RI FRXUVH GLPHQVLRQOHVV DV DUH DOO PHDVXUHV RI UHODWLYH FRQILJXUDWLRQ

PAGE 22

$OWHUQDWLYHO\ PRWLRQ ZULWWHQ LQ WHUPV RI WKH GLVSODFHPHQW YHFWRU X LV [ ; X;Wf RU [ L ; L X;L;; Wf f $QG WKH GHULYDWLYH RI X WKH GLVSODFHPHQW JUDGLHQW LV X ; RU DX L D[ M f ,I WZR SDUWLFOHV LQ WKH UHIHUHQFH FRQILJXUDWLRQ DUH DQ LQILQLn WHVLPDO GLVWDQFH G; DSDUW WKHQ WKHVH WZR SDUWLFOHV DUH DQ LQILQLWHVn LPDO GLVWDQFH G[ DSDUW LQ WKH GHIRUPHG FRQILJXUDWLRQ VHH )LJXUH f 7HQVRU ) PDSV WKH QHLJKERUKRRG RI SDUWLFOH 3 LQ WKH )LJXUH 7KH PRWLRQ RI WKH QHLJKERUKRRG RI SDUWLFOH 3

PAGE 23

UHIHUHQFH FRQILJXUDWLRQ WR LWV QHLJKERUKRRG LQ WKH GHIRUPHG FRQILJXUn DWLRQ 7KXV WKH OLQHDU WUDQVIRUPDWLRQ LV G[ )rG; RU G[ ) G; a a L LM M f 8QLTXH PDSSLQJ LV JXDUDQWHHG DV ORQJ DV WKH -DFRELDQ _)_ LV JUHDWHU WKDQ ]HUR DQG ILQLWH 6LQFH YHFWRU G; KDV OHQJWK G6 DQG YHFWRU G[ KDV OHQJWK GV /DJUDQJLDQ VWUDLQ ( PD\ EH GHILQHG LQ WHUPV RI WKHVH OHQJWKV VTXDUHG DV GVf G6f G;(G; f 7KH *UHHQ GHIRUPDWLRQ WHQVRU & UHIHUUHG WR WKH XQGHIRUPHG FRQ ILJXUDWLRQ JLYHV WKH QHZ VTXDUHG OHQJWK GVf RI WKH HOHPHQW LQWR ZKLFK WKH JLYHQ HOHPHQW G; LV GHIRUPHG 6\PEROLFDOO\ WKH UHODWLRQn VKLS LV f ZKHUH 7 & ) ) f &RQYHUVHO\ WKH &DXFK\ GHIRUPDWLRQ WHQVRU % JLYHV WKH LQLWLDO VTXDUHG OHQJWK G6f RI DQ HOHPHQW G[ LQ WKH GHIRUPHG

PAGE 24

FRQILJXUDWLRQ 7KLV UHODWLRQVKLS LV ZULWWHQ DV G6f G[% rG[ f ZKHUH 7 % ) f ) f /DJUDQJLDQ VWUDLQ ZULWWHQ LQ WHUPV RI WKH *UHHQ GHIRUPDWLRQ WHQVRU LV f ,Q WHUPV RI WKH GLVSODFHPHQW JUDGLHQW ( LV X X X X ( >f§ f§ f§ f§ LM /; ; ; ; L L f 6WUDLQ FRPSXWHG E\ D 7D\ORU VHULHV >@ LV ( f§ f§fL;fOfLf ; 9; 9; 9; ( f§ f§ff§ff§f < Y X < Y X X ; ; < Y Y ; < Z Z ;
PAGE 25

)RU PDQ\ HQJLQHHULQJ DSSOLFDWLRQV WKH VRFDOOHG VPDOO GLVn SODFHPHQW WKHRU\ LV DGHTXDWH DQG PXFK HDVLHU WR LPSOHPHQW %\ WKLV WKHRU\ VWUDLQ LV X e ; UX YQ H f§Uf§ f§L O< ;M f Y e f ; ,I WKH -DFRELDQ LV JUHDWHU WKDQ ]HUR DQG ILQLWH WKH GHIRUPDn WLRQ JUDGLHQW ZULWWHQ LQ WHUPV RI D URWDWLRQ WHQVRU 5 DQG WKH ULJKW VWUHWFK WHQVRU 8 D SRODU GHFRPSRVLWLRQ LV ) 58 f 7HQVRUV 5 DQG 8 DUH XQLTXH DQG UHSUHVHQW ULJLGERG\ URWDWLRQ DQG VWUHWFK UHVSHFWLYHO\ 5RWDWLRQ WHQVRU 5 LV RUWKRJRQDO DQG SURGXFHV D ULJLGERG\ URWDWLRQ EHWZHHQ WKH SULQFLSDO D[HV RI & DW ; DQG WKH SULQFLSDO D[HV RI % DW [ 7KH QHFHVVDU\ DQG VXIILFLHQW FRQGLWLRQ IRU QR ORFDO ULJLGERG\ URWDWLRQ VRPHWLPHV FDOOHG SXUH VWUDLQf GXULQJ PRWLRQ LV 5 6\PPHWULF SRVLWLYHGHILQLWH 8 SURGXFHV WKH FKDQJHV LQ YHFWRU OHQJWKV GXULQJ PRWLRQ DQG SURGXFHV URWDWLRQ LQ DGGLWLRQ WR 5 RI DOO YHFWRUV H[FHSW WKRVH LQ WKH SULQFLSDO

PAGE 26

GLUHFWLRQV RI 8 1RWH WKDW 8 LV QHFHVVDU\ DQG VXIILFLHQW WR GHILQH ORFDOO\ SXUH ULJLGERG\ QRWLRQ QHJOHFWLQJ WKH WULYLDO FDVH ZKHQ QR PRWLRQ RFFXUVf DQG WUDQVODWLRQ GRHV QRW FKDQJH D YHFWRU ZLWK UHVSHFW WR WKH FRPPRQ UHIHUHQFH D[LV 7HQVRUV 8 DQG 5 ZULWWHQ LQ WHUPV RI ) DQG & DUH 8 & f 5 )r8 f $V D ILQDO QRWH WKH GHIRUPDWLRQ DW DQ\ SRLQW PD\ UHVXOW IURP WUDQVODWLRQ D ULJLGERG\ URWDWLRQ RI WKH SULQFLSDO D[HV RI VWUDLQ DQG VWUHWFKHV DORQJ WKHVH D[HV 7KHVH PRWLRQV PD\ RFFXU LQ DQ\ RUGHU EXW WKHLU WHQVRULDO PHDVXUHV PD\ QRW 0DWKHPDWLFDOO\ /DJUDQJLDQ PRWLRQ LV DVVXPHG DV D VXFFHVVLYH DSSOLFDWLRQ RI VWUHWFK E\ 8 ULJLGERG\ URWDWLRQ E\ 5 WUDQVODWLRQ WR [ ,, ,PDJH 3URFHVVLQJ DQG 3DWWHUQ 5HFRJQLWLRQ ,, 'LJLWDO ,PDJHV $ W\SLFDO LPDJH SURFHVVLQJ V\VWHP LV VKRZQ LQ )LJXUH &RQYHUn VLRQ RI WKH LQWHQVLW\ GLVWULEXWLRQ RI D FDPHUD LPDJH WR D IRUP VXLWn DEOH IRU FRPSXWHUV GLJLWL]LQJ LV DFFRPSOLVKHG E\ WKH GLJLWL]HU $IWHU GLJLWL]DWLRQ WKH LPDJH LV WUDQVIHUUHG WR WKH

PAGE 27

FRPSXWHU IRU IXUWKHU SURFHVVLQJ +RZ DQ LPDJH LV SURFHVVHG GHSHQGV RI FRXUVH RQ WKH KDUGZDUH FRQILJXUDWLRQ >@ 2EMHFW 0LFURFRPSXWHU 'LJLWL]HU ZLWK ,QWHUQDO 0RQLWRU 'LJLWL]HU )LJXUH $ W\SLFDO 3&EDVHG LPDJH SURFHVVLQJ V\VWHP $ GLJLWDO LPDJH LV DQ DSSUR[LPDWLRQ RI D UHDO FRQWLQXRXV LPDJH )RU H[DPSOH WKH UHDO LPDJH RI WKH VSRW LQ )LJXUH D LV UHn SUHVHQWHG LQ GLJLWDO IRUP DV WKH PDWUL[ RI LQWHJHU QXPEHUV LQ )LJXUH E (DFK PDWUL[ HOHPHQW RU SL[HO ZKLFK LV GLPHQVLRQOHVVf FRUUHn VSRQGV WR D GLVFUHWH SRLQW LQ WKH UHDO LPDJH DQG WKH YDOXH RI WKH HOHPHQW JUD\ OHYHOf LV WKH DSSUR[LPDWLRQ RI WKH DYHUDJH ORFDO LOOXPn LQDWLRQ LQWHQVLW\ 1RWLFH WKDW EODFN KDV ORZHU JUD\ OHYHOV WKDQ

PAGE 28

ZKLWH DQG WKDW WKH HGJH UHJLRQ ERXQGDU\f EHWZHHQ WKH WZR KDV LQWHUn PHGLDWH YDOXHV 7KH YDULDWLRQ RI JUD\ OHYHOV LQ UHJLRQV RI FRQVWDQW LOOXPLQDWLRQ LV WKH HIIHFW RI QRLVH *UD\ OHYHOV UDQJH IURP ]HUR WR VRPH PD[LPXP ZKLFK GHSHQGV RQ WKH KDUGZDUH HJ FDPHUD GLJLWL]HUf 5HDO ,PDJH + r r r } r r r } r r r r r } LHR r r A r r r } r r r r r r r r r r r r r 'LJLWDO ,PDJH Df Ef )LJXUH $ VSRW DQG LWV GLJLWDO LPDJH 7KH YDOXHV ,[\f LQ WKH GLJLWDO LPDJH FRUUHVSRQG WR LOOXPLQDWLRQ LQWHQVLW\ LQ WKH UHDO LPDJH 1RLVH FRUUXSWV WKH LPDJH JUD\ OHYHOV DQG WKH VRXUFHV PD\ EH FODVVLILHG LQWR WZR JURXSV RSWLFDO DQG HOHFWURQLF 7ZR H[DPSOHV RI RSWLFDO QRLVH DUH SDUWLFXODWH PDWWHU LQ WKH DLU DQG ODVHU VSHFNOH (OHFWURQLF QRLVH FRPHV IURP WKH GLVFUHWL]DWLRQ SURFHVV WKHUPDO

PAGE 29

VRXUFHV LQ WKH KDUGZDUH LQWHUIHUHQFH IURP H[WUDQHRXV HOHFWURPDJQHWLF VRXUFHV HWF 1RLVH LV XVXDOO\ DVVXPHG VWDWLVWLFDOO\ UDQGRP DQG QRUPDOO\ GLVWULEXWHG ZLWK D ]HUR PHDQ S DQG D VWDQGDUG GHYLDn WLRQ D 8QIRUWXQDWHO\ QR NQRZQ PHWKRG FRPSOHWHO\ HOLPLQDWHV WKH Q HIIHFWV RI QRLVH (DFK DUUD\ HOHPHQW KDV QHLJKERUV ZKLFK DUH FRPPRQO\ FDOOHG WKH HLJKWQHLJKERUV >@ ,Q )LJXUH WKH FHQWHU HOHPHQW SL[HO [\f LV WKH UHIHUHQFH HOHPHQW RU HOHPHQW RI LQWHUHVW 7KH HOHPHQW GLUHFWO\ DERYH LW LV DUELWUDULO\ ODEHOHG QHLJKERU DQG WKH RWKHU HOHPHQWV DUH QDPHG LQ D FRXQWHUFORFNZLVH GLUHFWLRQ $OWKRXJK HOHn PHQWV DORQJ WKH LPDJHnV RXWHU DUUD\ HGJH GR QRW KDYH DOO 3L[HO [\f $ )LJXUH $ SL[HO DQG LWV HLJKWQHLJKERUV

PAGE 30

QHLJKERUV WKH QHLJKERUV WKH\ GR KDYH DUH DOVR ODEHOHG DFFRUGLQJ WR )LJXUH 6SDWLDO DQG LOOXPLQDWLRQ VDPSOLQJ PXVW EH VXIILFLHQW WR UHSUHn VHQW DGHTXDWHO\ WKH UHDO LPDJH 7RR ORZ D VSDWLDO VDPSOLQJ IUHTXHQF\ DYHUDJHV LQWHQVLW\ RYHU WRR ODUJH DQ DUHD VHH )LJXUH LQ UHIHUn HQFH f ZKLOH WRR KLJK D VDPSOLQJ IUHTXHQF\ UHVXOWV LQ WKH ORVV RI LQIRUPDWLRQ RXWVLGH WKH ILHOG RI YLHZ ,Q JHQHUDO LQFUHDVLQJ LQWHQn VLW\ FRUUHVSRQGV WR LQFUHDVHG JUD\ OHYHOV EXW LQWHQVLWLHV EH\RQG WKH FDSDELOLW\ RI WKH HTXLSPHQW DUH OLPLWHG WR WKH PD[LPXP JUD\ OHYHO $Q LQFUHDVH LQ WKH UDQJH RI JUD\ OHYHOV LQFUHDVHV LPDJH GHWDLO VHH )LJXUH LQ UHIHUHQFH f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n WLYHO\ VLPSOH WKH\ DUH HDVLO\ VHJPHQWHG +LVWRJUDPPLQJ 6HJPHQWDWLRQ XVXDOO\ EHJLQV ZLWK KLVWRJUDP PLQJ :KHQ WKH IUHTXHQF\ GLVWULEXWLRQ RI WKH JUD\ OHYHOV LQ DQ LPDJH

PAGE 31

LV FRPSXWHG IRU WKH HQWLUH LPDJH FRQWUDVW DQG WKH SUHVHQFH RI PXOWLn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

PAGE 32

WKH QHHGHG LQIRUPDWLRQ LV HDVLO\ HVWLPDWHG ZLWKRXW DFWXDOO\ FRPSXWLQJ D KLVWRJUDP 7KUHVKROGLQJ 7KUHVKROGLQJ IROORZV KLVWRJUDPPLQJ DQG LQFOXGHV WKH WDVN RI GHWHUPLQLQJ WKH JUD\ OHYHO ZKLFK LQ WKLV FDVH UHDVRQn DEO\ VHSDUDWHV UHJLRQV RI EODFN IURP ZKLWH 7KLV LV IUHTXHQWO\ DFFRPSOLVKHG LQWHUDFWLYHO\ RU E\ XVH RI D SULRUL LQIRUPDWLRQ DQG WKH FKRLFH RI WKH RSWLPXP WKUHVKROG LV D VXEMHFW RI FRQVLGHUDEOH LQWHUHVW >@ 7KH DOWHUQDWLYH DSSURDFK LQ WKLV VWXG\ UHTXLUHV QHLWKHU LQWHUn DFWLYH WKUHVKROGLQJ QRU DQ RSWLPXP WKUHVKROG ,QVWHDG D JRRG HVWLn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f ,Q WKH VHFRQG VWHS WKH VWDQGDUG GHYLDWLRQ LV VXEWUDFWHG IURP WKH DYHUDJH UHVXOWLQJ LQ ,J $OJHEUDLFDOO\ WKLV LV f

PAGE 33

&UD\ /HYHOV )LJXUH %ODFN DQG ZKLWH UHJLRQV LQ D ELPRGDO KLVWRJUDP A , ,W f $QRWKHU VHW RI RU PRUH SL[HOV ZKRVH YDOXHV DUH EHORZ ,J DUH UDQn GRPO\ VHOHFWHG DQG XVHG WR FRPSXWH WKH DYHUDJH A 7KXV WKH DYHUDJH EODFN OHYHO LV DSSUR[LPDWHG E\ A &KHE\FKHYnV WKHRUHP >@ VWDWHV WKDW LI D SUREDELOLW\ GLVWULEXn WLRQ KDV D PHDQ S DQG D VWDQGDUG GHYLDWLRQ D WKH SUREDELOLW\ RI REWDLQLQJ D YDOXH [ ZKLFK GHYLDWHV IURP WKH PHDQ E\ DW OHDVW VWDQGn DUG GHYLDWLRQV LV DW PRVW 0DWKHPDWLFDOO\ WKLV LV H[SUHVVHG DV c& .Rf A 3_[S f

PAGE 34

6LQFH RQO\ RQH VWDQGDUG GHYLDWLRQ LV VXEWUDFWHG IURP WKH DYHUDJH ILQGLQJ SL[HOV GDUNHU WKDQ ,J LV KLJKO\ SUREDEOH $ VNHZ RI WKH KLVWRJUDP WR WKH OHIW IXUWKHU LPSURYHV WKH RGGV RI ILQGLQJ SL[HOV OHVV WKDQ ,J 7KH WKLUG VWHS LV WKH LQLWLDO HVWLPDWLRQ RI WKH DYHUDJH ZKLWH JUD\ OHYHO $ VXIILFLHQW UDQGRP VDPSOH ZKRVH YDOXHV DUH JUHDWHU WKDQ ,$ LV XVHG WR FRPSXWH WKH LQLWLDO HVWLPDWH RI )LQDOO\ WKH WKUHVKROG YDOXH ,W LV FRPSXWHG XVLQJ WKH HVWLPDWHV IRU WKH EODFN DQG ZKLWH OHYHOV 6\PEROLFDOO\ ,PDJHV GHVLJQHG IRU WKLV VWXG\ DUH DVVXPHG WR KDYH WZR UHJLRQV EODFN DQG ZKLWH ,Q D VHQVH WKH\ DUH ELQDU\ UHJLRQf LPDJHV 7KXV VHJPHQWDWLRQ LV HDVLO\ DFFRPSOLVKHG E\ ORFDWLQJ DQG LVRODWLQJ WKH EODFN UHJLRQV RI LQWHUHVW IURP WKH ZKLWH EDFNJURXQG 5DVWHU VFDQQLQJ DQG ERUGHU IROORZLQJ %HJLQQLQJ DW WKH RULJLQ WKH XSSHU OHIW FRUQHU WKH LPDJH LV UDVWHU VFDQQHG LH DORQJ D URZ RU FROXPQf XQWLO D EODFN SL[HO LV IRXQG ,I DW OHDVW RQH RI WKH QHLJKERUV LV EODFN D SRVVLEOH EODFN UHJLRQ LV DVVXPHGVSXULRXV QRLVH LV DVVXPHG LI WKH SL[HO KDV QR EODFN QHLJKERUV ,I D EODFN QHLJKERU LV IRXQG WKH ERUGHU EHWZHHQ EODFN DQG ZKLWH LV IROORZHG DURXQG WKH EODFN UHJLRQ WKHUHE\ LVRODWLQJ LW IURP WKH QHLJKERULQJ ZKLWH

PAGE 35

5RVHQIHOG >@ KDV GHYHORSHG D JHQHUDOSXUSRVH ERUGHUIROORZLQJ DOJRULWKP )LUVW JUD\ OHYHOV DUH UHGXFHG E\ WKUHVKROGLQJ WR nV IRU EODFN RU nV IRU ZKLWH 7KLV LV TXLFNO\ GRQH LI RQH KDV WKH DSSURSULDWH KDUGZDUHf %\ FRPSDULQJ QHLJKERUV DQG ORFDWLQJ WKH QH[W HGJH SL[HO WKH DOJRULWKP IROORZV WKH ERUGHU FRXQWHUFORFNZLVH IRU DQ RXWHU ERUGHU DQG FORFNZLVH IRU DQ LQQHU RQH 6LQFH ERUGHUV PD\ RYHUODS SURYLVLRQV DUH PDGH IRU EDFNWUDFNLQJ ,Q WKLV VWXG\ WKH JHQHUDOLW\ RI 5RVHQIHUGnV DOJRULWKP LV XQn QHFHVVDU\ DQG WKH DOJRULWKP LV PRGLILHG IRU IDVWHU H[HFXWLRQ 7KH PRGLILHG DOJRULWKP VHH DSSHQGL[f LV GHVLJQHG WR IROORZ RXWHU HGJHV QR LQQHU HGJHV H[LVWf LQ WKH FORFNZLVH GLUHFWLRQ DQG UHGXFLQJ WKH LPDJH LV HOLPLQDWHG E\ FRPELQLQJ WKUHVKROGLQJ ZLWK ERUGHU IROORZLQJ 7KH RQO\ LQIRUPDWLRQ VDYHG GXULQJ ERUGHU IROORZLQJ LV WKH \ FRRUGLQDWH RI WKH VSRWnV OHIWPRVW SL[HO / WKH \ FRRUGLQDWH RI WKH ULJKWPRVW SL[HO 5 WKH [ FRRUGLQDWH RI WKH KLJKHVW SL[HO 7 DQG WKH [ FRRUGLQDWH RI WKH ORZHVW SL[HO % 6LQFH WKH EODFN UHJLRQ H[WHQGV D OLWWOH EH\RQG WKH DSSUR[LPDWH ERUGHU SL[HOV DUH VXEn WUDFWHG IURP / DQG 7 ZKLOH SL[HOV DUH DGGHG WR WKH 5 DQG % VHH )LJXUH f 7KXV WKH ORFDO UHJLRQ RI LQWHUHVW EODFN DQG LQWHUn PHGLDWH JUD\f LV IUDPHG IRU IXUWKHU SURFHVVLQJ 5DVWHU VFDQQLQJ UHVXPHV DW WKH ERWWRP RI WKH IUDPH DQG FRQWLQXHV XQWLO DQRWKHU EODFN UHJLRQ LV IRXQG DQG LW LV IUDPHG 6FDQQLQJ DQG IUDPLQJ FRQWLQXH XQWLO WKH HQWLUH LPDJH LV FRYHUHG 7KXV WKH LPDJH LV FRPSOHWHO\ VHJPHQWHG DQG WKH LQIRUPDWLRQ QHHGHG IRU IHDWXUH H[n WUDFWLRQ LV DYDLODEOH

PAGE 36

U< ; (GJH 7 f§ )UDPH n f n 8! / / ,62 (GJH % )LJXUH )UDPH RI VSRW LQ )LJXUH ,PDJH )HDWXUHV ,PDJHV FRQWDLQ YDULRXV W\SHV RI LQIRUPDWLRQ NQRZQ DV IHDWXUHV DQG ZKHQ SURSHUO\ VHOHFWHG WKH\ DGHTXDWHO\ UHSUHVHQW WKH LPDJHV +HUH LW LV DVVXPHG WKDW QDWXUDO IHDWXUHV DUH LQWULQVLF WR WKH UHDO VFHQH DQG DUWLIDFWV DUH DGGHG DV GHVLUHG WR DLG VXEVHTXHQW SURFHVn VLQJ )HDWXUHV ERWK QDWXUDO DQG DUWLIDFW DUH JHQHUDOO\ JURXSHG LQWR WKUHH FODVVHV SK\VLFDO VWUXFWXUDO DQG PDWKHPDWLFDO >@

PAGE 37

3K\VLFDO DQG VWUXFWXUDO IHDWXUHV DUH LQ WKLV FDVH WKH DUWLn IDFWV VKRZQ LQ )LJXUH SOXV WKH SDWWHUQ WKH\ IRUP LQ WKH LPDJH )RU VLPSOLFLW\ WKH VLQJOHf LFRQ LV FRQVLGHUHG D VSHFLDO VSRW DQG DOO RWKHU VSRWV DUH DVVXPHG XQLIRUP LQ VL]H 6SRW ,FRQ )LJXUH ,PDJH DUWLIDFWV 7KH LFRQ LV D VSHFLDO VSRW 0DWKHPDWLFDO IHDWXUHV DUH IUDPH DUHDV VSRW FHQWURLGV DQG WKRVH FRQFHUQLQJ LFRQ LGHQWLILFDWLRQ DQG URWDWLRQ 7KHVH IHDWXUHV DOORZ H[WHQVLYH LQWHJHU SURJUDPPLQJ ZKLFK VSHHGV WKH PDMRULW\ RI WKH FDOn FXODWLRQV DQG LPSURYHV DFFXUDF\

PAGE 38

,, )HDWXUH ([WUDFWLRQ +HUH IHDWXUH H[WUDFWLRQ FRQVLVWV RI LVRODWLQJ WKH DUWLIDFWV DQG FRPSXWLQJ WKHLU GHVLUHG PDWKHPDWLFDO IHDWXUHV 6LQFH WKH DUWLIDFWV DUH H[WUDFWHG E\ IUDPLQJf GXULQJ VHJPHQWDWLRQ DOO WKDW UHPDLQV LV WKH FRPSXWDWLRQ )UDPH DUHD )UDPH LQWHJHUf DUHD LV HDVLO\ FRPSXWHG XVLQJ WKH LQIRUPDWLRQ REWDLQHG GXULQJ WKH VHJPHQWDWLRQ SURFHVV 7KH FRPSXWDn WLRQ LV $ 5/fr%7f f 6LQFH WKH LFRQnV IUDPH DUHD LV PXFK ODUJHU WKDQ WKH RWKHU VSRWVn FODVVLI\LQJ HDFK DUWLIDFW DV VSRW RU LFRQ LV D VLPSOH ELQDU\ WDVN 6SRW FHQWURLGV 7KH FHQWURLGV RI DOO VSRWV DUH XVHG DV GHILQLn WLRQV RI VSRW SRVLWLRQ ,I WKH JUD\ OHYHOV LQ D VSRWnV IUDPH DUH DVVXPHG WR EH ZHLJKWV WKH [ DQG \ FRRUGLQDWHV RI WKH FHQWURLG DUH FDOFXODWHG >@ E\ % 5 = = [O[\f \ B ; 7 \ / % 5 = = ,[\f [ 7 \ / f 5 % = = \O[\f \ \ / ; 7 5 % = = ,[\f \ / [ 7

PAGE 39

+HQFHIRUWK WKHVH VXPPDWLRQ OLPLWV DUH LPSOLHG XQOHVV RWKHUZLVH QRWHG +RZ GR JUD\ OHYHOV QRLVH DQG VSRW VL]H DIIHFW VSDWLDO UHVROXn WLRQ RI WKH FHQWURLG" +HUH LW LV HYLGHQW VHH )LJXUH f WKDW PHUHO\ VXPPLQJ ZLWK JUD\ OHYHOV ,[\f GRHV QRW \LHOG WKH SURSHU UHVXOW EHFDXVH WKH ZKLWH SL[HOV VKLIW WKH VSRWnV FRPSXWHG FHQWURLG DZD\ IURP WKH FRUUHFW YDOXH $ WUDQVIRUPDWLRQ RI DOO JUD\ OHYHOV LQ WKH IUDPH E\ WKH HTXDWLRQ ,n[\f ,[\f f Z ZKHUH LV WKH ORFDO DYHUDJH ZKLWH OHYHO DQG ,n[\f LV WKH QRLV\ JUD\ OHYHO GLIIHUHQFH EHWZHHQ EODFN DQG ZKLWHf JLYHV WKH DSSUR[Ln PDWH FHQWURLG FRRUGLQDWHV $QG VLQFH SL[HOV RQ WKH IUDPH HGJH DUH ZKLWH WKHLU DYHUDJH JLYHV D JRRG ORFDO HVWLPDWH IRU ,Z 7KH VXEn VWLWXWLRQ RI ,n[\f LQWR HTXDWLRQV f OHDGV WR = ( [ ,n[\f Y /B] ( ( ,n[\f [ \ f ( ( \ ,n[\f \ \ [ e ,n [ \f \ [

PAGE 40

7R DFFRXQW IRU QRLVH DVVXPH HDFK JUD\ OHYHO LQ WKH LPDJH FRQn VLVWV RI WKH DFWXDO LH XQFRUUXSWHG E\ QRLVHf YDOXH U[\f SOXV ]HURPHDQ *DXVVLDQ QRLVH Q[\f 7KXV ,n[\f U[\f Q[\f Z ,n[\f +[\f Q[\f f ZKHUH +[\f 7[\f LV WKH DFWXDO JUD\ OHYHO GLIIHUHQFH Z 6XEVWLWXWLQJ HTXDWLRQ f LQWR HTXDWLRQV f OHDGV WR (([+ (([Q [\ [\ ((+ ((Q [\ [\ f ((\+ ((\Q \ \[ <[ ((+ ((Q \[ \[ ,I WKH H[SHFWHG YDOXH >@ IRU WKH QRLVH LV DVVXPHG (^Q[\f` ((Q[\f [\ r ((Q[\f \[ f (Q[\f [ (Q[\f \

PAGE 41

DQG VXEVWLWXWHG LQWR HTXDWLRQV f WKHQ WKH H[SHFWHG FHQWURLG LV (^;` H^[` H^\` (([+ ([(T [\ ; \ ((+ ((T [\ [\ (([+ (^Q`([ [\ [ ((+ (^Q` [\ ((\+ (^Q`(\ \[ \ ((+ 6LQ \[ f 6LQFH (^Q` HTXDWLRQV f EHFRPH (^;` (([+ [\ ((+ [\ (_<` ((\+ \[ ((+ \[ f ,Q JHQHUDO WKH VXPPDWLRQV LQ HTXDWLRQV f DUH QRW HTXDO WR ]HUR ZKLFK LQWURGXFHV UDQGRP HUURU %XW LI (T WKHQ ; (^;` \ DQG LI (T WKHQ < (^<` 7KXV WKH VXPPDWLRQ SURFHVV XVHG WR [ FRPSXWH WKH FHQWURLGV JUHDWO\ UHGXFHV DOWKRXJK LQ JHQHUDO LW GRHV QRW HOLPLQDWH WKH HIIHFW RI QRLVH

PAGE 42

7R XQGHUVWDQG EHWWHU KRZ QRLVH VSRW GLDPHWHU DQG QXPEHU RI JUD\ OHYHOV LQWHUDFW WR DIIHFW VSDWLDO UHVROXWLRQ FRQVLGHU WKH RQH GLPHQVLRQ PRGHO LQ )LJXUH 7KLV PRGHO LV HVVHQWLDOO\ D FROXPQ LQ WKH WZRGLPHQVLRQDO LPDJH VHH )LJXUH f ZLWK IUDPH HGJHV DW [ 7 DQG [ % )RU VLPSOLFLW\ HGJH SL[HOV DUH DOZD\V ORFDWHG DW [ 7 DQG [ % 7KH H[WHQVLRQ WR WKH JHQHUDO WZRGLPHQVLRQ FDVH LV VWUDLJKWn IRUZDUGf 6LPLODU WR HTXDWLRQ f WKH FHQWURLG FDOFXODWLRQ LV =[+[f ([T[f (+[f (Q[f ; ; f )UDPH )LJXUH 2QHGLPHQVLRQ PRGHO $VVXPH

PAGE 43

[ 7 [ % 7[% RWKHUZLVH + + Ff§ P .B 7 rf§%f§ DQG 7% ZKHUH +F D FRQVWDQW RYHU WKH OLQH OHQJWK LV WKH JUD\ OHYHO GLIIHUn HQFH DQG +P LV WKH PD[LPXP SRVVLEOH JUD\ OHYHO GLIIHUHQFH FRQWUDVWf EHWZHHQ EODFN DQG ZKLWH 7KH LQIOXHQFH RI QRLVH DQG WKH HGJH SL[HOV FDQ EH GHVFULEHG LQ WHUPV RI UHVXOWDQWV 7KH QRLVH UHVXOWDQW QF LV WKH SURGXFW RI UDQGRP YDULDEOHV DQG Q LH ([Q[f Q DQG (Q[f Q f 'LPHQ F F F ; ; VLRQOHVV SDUDPHWHU LV WKH IUDFWLRQ RI WKH [ 7 HGJH SL[HO FRYHUHG E\ WKH OLQH DQG N + LV WKH UHVXOWDQW $VVXPH WKH VDPH IRU N & DQG .Q+ DW HGJH [ % 'XULQJ GLJLWL]DWLRQ NB DQG NG DUH GLVFUH & ,' WL]HG VHH )LJXUH f WR G DQG UHVSHFWLYHO\ )RU D JLYHQ OLQH L VHJPHQW DW D VSHFLILHG ORFDWLRQ DQG DUH FRQVWDQWV 7KH VXE VWLWXWLRQ RI WKHVH YDOXHV LQWR HTXDWLRQ f \LHOGV % ( [+ + >N 7f%f@Q B [ 7 F F 7 % r [ L+;f+F9A%fULF % ( [ 7f7f7f %f%f%f [ 7 .7+ N+ % F +[f ,I

PAGE 44

LV DGGHG WR ( [ %f%f%f 7f7f7f [ 7 WKHQ % ( [ 7%f7%f7%f 7%f7%f7%f [ 7 DQG % ( [ _7%f [ 7 ZKHUH Q LV WKH QXPEHU RI WHUPV LQ WKH VXP 6LQFH %B!7 DQG Q! Q %f 7f DQG Q %7 7KHUHIRUH % ( [ [ 7 %7f7%f % ( +[f %7f+ [ 7 DQG

PAGE 45

6XEVHTXHQWO\ HTXDWLRQ f EHFRPHV %7f7r%f+Fr_6QF}+F877rfrF%f@f >%7f+r9+F9NEf@ a $IWHU GLYLGLQJ WKH WRS DQG ERWWRP E\ + HTXDWLRQ f EHFRPHV %7af7%f> A77f%%f@ >%7f&WE f ZKHUH e f§ 1RWH WKH e UDWLR LV VLPLODU WR WKH LQYHUVH RI K & 3UDWWnV >@ VLJQDOWRQRLVH UDWLR + 615 f§f D f )UDFWLRQ RI (GJH 3L[HO FRYHUHG )LJXUH 7KH GLVFUHWL]DWLRQ RI DQG LF $Q LQFUHDVH LQ +F LPn SURYHV VSDWLDO UHVROXWLRQ RI WKH HGJHV

PAGE 46

$IWHU VWXG\LQJ )LJXUH DQG HTXDWLRQ f LW WKDW 7 RU % JLYHV WKH PRVW QRLVH HUURU ZKLOH %7f7%f>N7fN%f@ ; B L 6 >%7f7%O JLYHV QRQH ,I D OLQH OHQJWK RI G %7a LV DVVXPHG HTXDWLRQ f GfG7f>eNW7fN G7f@ Y / >GLF N @ ,I GfG7f>NW7fBG7f@ ; >G7.J@ ZKLFK LV WKH H[DFW FHQWURLG ,I 7 LV DVVXPHG DQG N VXEVWLWXWHG LQWR HTXDWLRQ f WKH UHVROXWLRQ HUURU LV GfGf&7.%Gf@ GfGf>AA.G 5H >G.7.@ >G&_A@ 7KH JHQHUDO UHODWLRQVKLS EHWZHHQ e G DQG 5J LV HTXDWLRQ f LV VLPSOLILHG 7R DVVXPH WKH OLQH SRVLWLRQ YDULDEOH LPSOLHV WKH IROORZLQJ H[SHFWHG YDOXHV H^NW` (^%` H^Nf` (^%` LV REVHUYHG f EHFRPHV f f n DQG n DUH U G f f§ f FOHDUHU LI LV D UDQGRP $IWHU VRPH DOJHEUD HTXDWLRQ f EHFRPHV

PAGE 47

B eGf H Gf f )RU e PXFK VPDOOHU WKDQ G HTXDWLRQ f VXJJHVWV DQ DOPRVW OLQHDU UHODWLRQ EHWZHHQ DQG 5J $ VSDWLDO GHSHQGHQF\ SHUKDSV GLIILFXOW WR YHULI\ H[SHULPHQWDOO\f LV DOVR LQGLFDWHG )LJXUH ,FRQ URWDWLRQ PHDVXUHG IURP WKH YHUWLFDO ,FRQ URWDWLRQ 1H[W WKH LFRQnV DQJOH RI URWDWLRQ LV FRPSXWHG VHH )LJXUH f XVLQJ WKH VHFRQG PRPHQW RI LQHUWLD 7KH DQDO\VLV LV DV IROORZV

PAGE 48

, (( ,n [\f G [[ \ [\ ,( ,n[\f G \\ ff [ f 3 ,( ,n[\f G G ;\ [\ ; \ ZKHUH ,[[ PRPHQW RI LQHUWLD DERXW LFRQnV [D[LV PRPHQW RI LQHUWLD DERXW LFRQnV \D[LV 3<9 SURGXFW RI LQHUWLD DERXW LFRQnV FHQWURLG G[ [ GLVWDQFH IURP SL[HO WR LFRQ FHQWURLG G\ \ GLVWDQFH IURP SL[HO WR LFRQ FHQWURLG ,FRQ URWDWLRQ LV 3 H 7DQUDf \\ [[ f 7KH DERYH HTXDWLRQ JLYHV WKH DQJOH WR ZLWKLQ D FRQVWDQW )RU H[DPSOH WKH DQDO\VLV JLYHV WKH VDPH DQJOH IRU ERWK LFRQ RULHQWDWLRQV LQ )LJXUH r 7KH GLIIHUHQFH LV UHVROYHG E\ FRQVWUXFWLQJ OLQH $$ WKURXJK WKH LFRQ FHQWURLG DQG FRPSXWLQJ WKH FHQWURLGV RI WKH XSSHU DQG ORZHU VHJPHQWV /LQH $$ KDV WKH IRUP \ P[ E $ $ ZKHUH [A DQG \A DUH SRLQWV RQ $$ ZLWK P 7$1Hf DQG E \7 P[ ,FRQ ,FRQ

PAGE 49

< )LJXUH ,FRQ URWDWLRQ LV GHWHUPLQHG WR ZLWKLQ D FRQVWDQW E\ HTXDn WLRQ f 7KH FHQWURLG IRU WKH XSSHU VHJPHQW LV ;$ = [ 7 \$ = \ / [Of[\f ;$ \$ = = ,n[\f [ 7 \ / ;$ \$ = = \Of[\f [ 7 \ / ;$ \$ = = ,n[\f [ 7 \ / f

PAGE 50

DQG IRU WKH ORZHU VHJPHQW ; 5 ( ( [On [ \f [ [$ \ \$ / % 5 ( ( ,n[\f [ [$ \ \$ f
PAGE 51

,QIRUPDWLRQ VDYHG GXULQJ IHDWXUH H[WUDFWLRQ LV RI FRXUVH IUDPH DUHD WKH FHQWURLG FRRUGLQDWHV RI HDFK VSRW WKH QXPEHU RI VSRWV ZKLFK VSRW LV WKH LFRQ DQG LFRQ URWDWLRQ 7KH DUHDV DQG WKH FHQn WURLGV DUH VWRUHG DV OLQHDU DUUD\V LQ WKH RUGHU WKH VSRWV ZHUH IRXQG /DVWO\ WKH SUREOHP RI VRUWLQJ WKLV UDZ LQIRUPDWLRQ DQG XVLQJ LW WR PDWFK VSRWV LQ WKH EHIRUH DQG DIWHUPRWLRQ LPDJHV LV VROYHG XVLQJ SDWWHUQ UHFRJQLWLRQ SULQFLSOHV ,, .QRZQ 3DWWHUQV ,Q WKLV VWXG\ SDWWHUQV DUH FODVVLILHG DV HLWKHU NQRZQ RU XQNQRZQ .QRZQ SDWWHUQV FRQWDLQ VWUXFWXUDO IHDWXUHV QDWXUDO DUWLIDFW RU ERWKf WKDW DUH UHFRJQL]DEOH EHIRUH DQG DIWHU PRWLRQ 8QNQRZQ SDWWHUQV DUH SDWWHUQV ZKLFK QHYHU KDYH UHFRJQL]DEOH VWUXFWXUH RU WKRVH LQ ZKLFK IHDWXUHV DUH REVFXUHG EH\RQG UHFRJQLWLRQ .QRZQ SDWWHUQV FRQWDLQLQJ RQO\ DUWLIDFWV DUH FRQVLGHUHG KHUH $UWLIDFWV DQG WKHLU DUUDQJHPHQW DUH VHOHFWHG WR LPSURYH VHJPHQn WDWLRQ IHDWXUH H[WUDFWLRQ PDSSLQJ DQG VXEVHTXHQW QXPHULFDO DQDO\VLV ([DPSOHV RI VLPSOH HDVLO\ DQDO\]HG SDWWHUQV DUH VKRZQ LQ )LJXUH 7KHVH SDWWHUQV DUH ILQH IRU VPDOO URWDWLRQV DQG DUH HDVLO\ UHFRJQL]HG EHFDXVH WKH UHODWLYH RUGHU RI WKH VSRWV GRHV QRW FKDQJH HJ WKH XSSHU OHIW VSRW UHPDLQV LQ WKH VDPH SRVLWLRQ UHODWLYH WR WKH RWKHU VSRWVf§GXULQJ PRWLRQ ,I URWDWLRQ LV GHJUHHV RU PRUH IRU WKH VTXDUH SDWWHUQV DQG GHJUHHV RU PRUH IRU WKH RWKHUV WKH UHODWLYH RUGHU LV ORVW 7KH SDWWHUQV LQ )LJXUH DOORZ VSRW LGHQWLILFDWLRQ DIWHU URWDWLRQV XS WR RQH UHYROXWLRQ 2QFH WKH SDWWHUQV LQ WKH XQGHIRUPHG DQG GHIRUPHG LPDJHV KDYH EHHQ

PAGE 52

)LJXUH 3DWWHUQV WKDW DUH HDVLO\ UHFRJQL]HG DQG DQDO\]HG )UDPHV DUH IRU UHIHUHQFH DQG QRW SDUW RI WKH SDWWHUQ )LJXUH 3DWWHUQV RI )LJXUH ZLWK LFRQV 7KH LFRQV DUH ORFDWHG IRU FRQYHQLHQFH

PAGE 53

UHFRJQL]HG VXEVHTXHQW DQDO\VLV LV VLPSOLILHG HVSHFLDOO\ IRU WKH /DJUDQJLDQ GHVFULSWLRQf EHFDXVH WKH XQGHIRUPHG VSDFLQJ LV DSSUR[Ln PDWHO\ FRQVWDQW ,, 6\QWDFWLF 3DWWHUQ 5HFRJQLWLRQ 3DWWHUQ UHFRJQLWLRQ LV XVXDOO\ GLYLGHG LQWR WZR FODVVHV GHFLn VLRQWKHRUHWLF DQG V\QWDFWLF ,Q WKH ILUVW FODVV D SDWWHUQ LV UHSUHVHQWHG E\ IHDWXUH YHFWRUV DQG UHFRJQL]HG E\ D VWRFKDVWLF SDUWLn WLRQLQJ RI IHDWXUH VSDFH >f§@ 6\QWDFWLF UHFRJQLWLRQ XVHV SDWWHUQ VWUXFWXUH DQG OLQJXLVWLF UXOHV IRU DQDO\VLV >@ 6\QWDFWLF UHFRJQLWLRQ DVVXPHV D SDWWHUQ LV GHVFULEHG E\ D ODQJXDJH ZKLFK FRQWDLQV YDOLG VHQWHQFHV FRPSRVHG IURP D ILQLWH DOSKDn EHW RU VHW RI V\PEROV 7KH ODQJXDJH LV JRYHUQHG E\ SDWWHUQ JUDPPDU RU VHW RI JUDPPDU UXOHV ,Q FRQWH[WIUHH ODQJXDJHV SDWWHUQ JUDPPDU LV XQUHVWULFWHG &RQWH[WVHQVLWLYH ODQJXDJHV UHVWULFW SDWWHUQ JUDPPDU RQ WKH QXPEHU RI VHQWHQFHV VHQWHQFH FRQVWUXFWLRQ RU ERWK +HUH WKH SDWWHUQ ODQJXDJH LV FRQWH[WVHQVLWLYH DQG KDV RQO\ RQH VHQWHQFH ZKLFK LV FRPSRVHG RI D IL[HG QXPEHU RI RUGHUHG V\PEROV 7KH DOSKDEHW V\PEROV DUH WKH VSRWVn UHODWLYH SRVLWLRQV ZKRVH RUGHU LV GHWHUPLQHG LQ WKH XQGHIRUPHG LPDJH $ VSRWnV FRQWH[W LV GHWHUPLQHG E\ LWV RU DOWHUQDWLYHO\ E\ WKURXJK f QHLJKERUV )RU H[DPSOH f KDV QHLJKERUV f f f DQG f $Q DWWHPSW WR PDS f LQ WKH XQGHIRUPHG LPDJH WR VD\ f LQ WKH GHIRUPHG LV RXW RI FRQWH[W ZLWK WKH QHLJKERUV VHH )LJXUH f 7KXV FRUUHFW SDWWHUQ PDSSLQJ LV DVVXUHG LI HYHU\ VSRW PDSV LQ FRQn WH[W ZLWK LWV QHLJKERUV

PAGE 54

f f f f f f f f f f f f f f f f f f f f f f f f 3DWWHUQ 2QHVHQWHQFH /DQJXDJH )LJXUH $ SDWWHUQ DQG LWV FRQWH[WVHQVLWLYH ODQJXDJH 1RWH WKH DOSKDEHW V\PEROVf DQG FRQWH[W LQ WKH ODQJXDJH +LJKHU GLPHQVLRQDO SDWWHUQ JUDPPDUV >a@ SURYLGH WKH FDSDELOn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

PAGE 55

DGGUHVVHV ,I ERWK LPDJHV DUH FDUHIXOO\ GLYLGHG LQWR D QXPEHU RI DGGUHVVHV VXEUHJLRQVf WKHQ D VSRW KDV WKH VDPH DGGUHVV EHIRUH DQG DIWHU PRWLRQRU QHDUO\ VR 7KH GHWHUPLQDWLRQ RI D VSRWnV DGGUHVV LV D VLPSOH FDOFXODWLRQ LQYROYLQJ FHQWURLG FRRUGLQDWHV DQG UHFRJQLWLRQ LV UHGXFHG WR PHUHO\ DUUDQJLQJ WKH VSRW DGGUHVVHV LQ SURSHU RUGHU 7KXV FRQWH[W LV UHFRJQL]HG 2IIVHWV DUH WKH VSRW DGGUHVVHV UHODWLYH WR WKH LFRQ DGGUHVV $OO RIIVHWV DUH GHILQHG DFFRUGLQJ WR DQ LFRQ DGGUHVV RI VHH )LJXUH f ,PDJH %RUGHU 6SRW $GGUHVVHV )LJXUH ,PDJH DGGUHVVHV DQG RIIVHWV LPDJH HGJHV 6RPH DGGUHVVHV RYHUODS

PAGE 56

3DWWHUQ JHRPHWU\ DIIHFWV WKH XSSHU OLPLWV RI PHDVXUDEOH VWUDLQ EHFDXVH ODUJH VWUDLQ GLVSODFHPHQWV PD\ FKDQJH WKH RUGHU RI WKH RIIn VHWV LQ WKH GHIRUPHG LPDJH 6SUHDGLQJ VSRWV IXUWKHU DSDUW LQ WKH XQGHIRUPHG LPDJHf LQFUHDVHV WKH VL]H RI D VSRWnV VXEUHJLRQ $ ODUJHU VXEUHJLRQ DOORZV WKH VSRW PRUH VWUDLQ GLVSODFHPHQW EHIRUH DQ\ FKDQJH LQ RIIVHW RUGHU RFFXUV 7KH GHWDLOV DUH EHVW LOOXVWUDWHG E\ H[DPSOH VHH )LJXUH f ,W LV DVVXPHG WKDW WKH XQGHIRUUDHG LPDJH LV DQDO\]HG ILUVW DQG ZLWK RQH H[FHSWLRQ ERWK LPDJHV UHFHLYH WKH VDPH SUHOLPLQDU\ DQDO\VLV L 8QGHIRUPHG )LJXUH $ VLPSOH H[DPSOH RI PRWLRQ X; X Y SL[HOV r ( f 7KH RULJLQ ALV LQ WKH XSSHU OHIW FRUQHU RI HDFK LPDJH

PAGE 57

6XFFHVVIXO PDSSLQJ LV LQLWLDOO\ DVVXPHG DQG ODWHU YHULILHG E\ IRXU ILQDO FKHFNV )LUVW WKH LPDJH LV VHJPHQWHG DQG WKH IHDWXUHV DUH H[WUDFWHG 6FDQQLQJ LV E\ FROXPQ DQG WKH VSRW FHQWURLGV DUH VWRUHG LQ WKH RUGHU WKH\ ZHUH IRXQG 6HH FROXPQV PDUNHG 5DZ LQ 7DEOHV DQG f 7KH LFRQ LV LGHQWLILHG DQG LWnV URWDWLRQ LV DOVR FRPSXWHG 7DEOH 8QRUGHUHG VSRW FRRUGLQDWHV LQ WKH XQGHIRUPHG LPDJH 5DZ 5HIHUHQFH )URP ,FRQ ; < ; < ; < 127( 8QLWV DUH SL[HOV 5DZ FRRUGLQDWHV DUH WKH VSRW FHQWURLGV LQ WKH LPDJH 5HIHUHQFH DQG )URP ,FRQ YDOXHV DUH FRPSXWHG E\ HTXDWLRQV f DQG f UHVSHFWLYHO\ 6HH )LJXUH

PAGE 58

7DEOH 8QRUGHUHG VSRW FRRUGLQDWHV LQ WKH GHIRUPHG LPDJH 5DZ 5HI HUHQFH )URP ,FRQ ; < ; < ; < 127( 8QLWV DUH SL[HOV 5DZ FRRUGLQDWHV DUH WKH VSRW FHQWURLGV LQ WKH LPDJH 5HIHUHQFH DQG )URP ,FRQ YDOXHV DUH FRPSXWHG E\ HTXDWLRQV f DQG f UHVSHFWLYHO\ 6HH )LJXUH $OO VSRWV DUH PRYHG LQWR D UHIHUHQFH SRVLWLRQ E\ WUDQVODWLRQ DQG URWDWLRQ VHH )LJXUH f 7UDQVODWLRQ LV DFFRPSOLVKHG E\ DGGLQJ ; WR WKH ; DQG
PAGE 59

ZKHUH ; DQG < DUH WKH FRRUGLQDWHV RI WKH LFRQ LQ UHIHUHQFH SRVL FS FS ) WLRQ W\SLFDOO\ WKH LPDJH FHQWHUf 5RWDWLRQ IRU HDFK VSRW LV FRPSXWHG IURP WKH [ DQG \ GLVWDQFH EHWZHHQ LW DQG WKH LFRQ 7KH FRPSXWDWLRQ LV [ [[7 S ,FRQ < << ,FRQ 9 ; < 3 3 7DQ Af ;5HI ;S>FRVXf]fFRV D!]@;7 <5HI VLQMM]fVLQ Z]@<7 f ZKHUH ;5JA DQG <5HA! DUH WKH QHZ FRRUGLQDWHV RI HDFK VSRW LQ WKH UHIHUHQFH SRVLWLRQ 7KH UHVXOWV RI WUDQVODWLRQ DQG URWDWLRQ DUH VKRZQ LQ 7DEOHV DQG XQGHU WKH 5HIHUHQFH FROXPQ 3RVLWLRQ UHODWLYH WR WKH LFRQ LH GLVWDQFH IURP WKH LFRQ LV [ [ < 5HO 5HI ,FRQ < < < 5HO 5HI ,FRQ f

PAGE 60

)LJXUH 7KH GHIRUPHG LPDJH DQG LWV UHIHUHQFH SRVLWLRQ 7KH GLVWDQFHV IURP WKH LFRQ DUH VKRZQ LQ 7DEOHV DQG XQGHU WKH FROXPQV PDUNHG )URP ,FRQ $OWKRXJK DOO VSRWV KDYH EHHQ ORFDWHG UHODWLYH WR WKH LFRQ UHFRJQL]LQJ WKH SDWWHUQ WKH\ IRUP DQG DUUDQJLQJ WKHP LQ SURSHU RUGHU E\ DGGUHVVf LV QRW DOZD\V HDV\HVSHFLDOO\ LI URWDWLRQ LV PRUH WKDQ D IHZ GHJUHHV $GGUHVV FRPSXWDWLRQ LV EDVHG RQ OHDUQLQJ WKH XQGHIRUPHG LPDJH 7KH VSRWV QHDUHVW DQG IDUWKHVW IURP WKH LFRQ URZ DQG FROXPQ DUH ORFDWHG LQ WKH XQGHIRUPHG LPDJH DQG XVHG WR FDOFXODWH WKH DGGUHVV RIIVHW $IIVHW E\ WKH IROORZLQJ

PAGE 61

ZKHUH ; 0$;;f PD[ ; 0,1;f UDP < 0$;
PAGE 62

1\ QXPEHU RI FROXPQV OHIW RU ULJKW RI LFRQ 1U:V QXPEHU RI URZV LQ WKH LPDJH ;JSDFV DSSUR[LPDWH URZ VSDFLQJ <6SDFH DSSUR[LPDWH FROXPQ VSDFLQJ 1JKLIW QXPEHU RI SRVVLEOH FROXPQV LQ LPDJH n)XQFWLRQ 1,17 URXQGV QXPEHUV WR WKH QHDUHVW LQWHJHU YDOXH 9DOXHV ;VSDFHf @ WKH RIIVHWV LQ DVFHQGLQJ RUGHU %\ HTXDWLRQ f RIIVHWV VHH FROXPQ 2IIVHW LQ 7DEOHV DQG f LQFUHDVH GRZQ FROXPQV DQG OHIW WR ULJKW &ROXPQ .H\ LV WKH RUGHU LQ ZKLFK WKH VSRWV ZHUH IRXQG 1RWH WKH WZRGLPHQVLRQ SDWWHUQ FRQWH[W LV QRZ UHSUHVHQWHG E\ D RQHGLPHQVLRQ DUUD\ ; ZKLFK FRUUHVSRQGV WR WKH RUGHUHG VSRW DGGUHVVHV 5HFRJQLWLRQ RI WKH WZRGLPHQVLRQ FRQWH[W LV DFFRPSOLVKHG E\ GHWHUPLQLQJ WKH URZFROXPQ RUGHU RI WKH VSRWV 7KH OHIWPRVW FROXPQ LV DVVXPHG WR EHJLQ ZLWK WKH ORZHVW RIIVHW DQG FRQWLQXH ZLWK DVFHQGn LQJ RIIVHW XQWLO WKH ; FRRUGLQDWH GHFUHDVHV &ROXPQ EHJLQV ZLWK WKH QH[W KLJKHVW RIIVHW DQG FRQWLQXHV XQWLO WKH ; FRRUGLQDWH DJDLQ GHFUHDVHV 7KH SURFHVV FRQWLQXHV XQWLO WKH RIIVHW DUUD\ LV H[KDXVWn HG %\ LQFUHPHQWLQJ URZ DQG FROXPQ FRXQWHUV WKH UHODWLYH VSRW RUGHU LV GHWHUPLQHG ZKLFK HVWDEOLVKHV FRQWH[W VHH 7DEOHV f 8QWLO QRZ ERWK XQGHIRUPHG DQG GHIRUPHG LPDJHV ZHUH SURFHVVHG WKH VDPH H[FHSW IRU OHDUQLQJ WKH XQGHIRUPHG LPDJHf DQG DVVXPHG WR KDYH WKH VDPH FRQWH[W 7KLV DVVXPSWLRQ LV YHULILHG E\ WKH IROORZLQJ

PAGE 63

7DEOH 2UGHUHG VSRW FRRUGLQDW HV LQ WKH XQGHIRUPHG LPDJH 5HIHUHQFH )URP ,FRQ 2IIVHW .H\ ; < ; < 127( 8QLWV DUH SL[HOV 2IIVHWV DUH FRPSXWHG E\ HTXDWLRQ f .H\V DUH WKH RUGHU LQ ZKLFK WKH VSRWV ZHUH IRXQG 8QGHIRUPHG VSDFLQJ LV SL[HOV LQ [ DQG \ 6HH )LJXUH

PAGE 64

7DEOH 2UGHUHG VSRW FRRUGLQDWHV LQ WKH GHIRUPHG LPDJH 5HIHUHQFH )URP ,FRQ 2IIVHW .H\ ; < ; < 127( 8QLWV DUH SL[HOV 2IIVHWV DUH FRPSXWHG E\ HTXDWLRQ f .H\V DUH WKH RUGHU LQ ZKLFK WKH VSRWV ZHUH IRXQG 6HH )LJXUH

PAGE 65

7DEOH 6SRW GLVWDQFH IURP RULJLQ LQ XQGHIRUPHG LPDJH &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < 127( 8QLWV DUH SL[HOV 7KH RULJLQ LV WKH XSSHU OHIW FRUQHU RI WKH LPDJH 6HH )LJXUH

PAGE 66

7DEOH 6SRW GLVWDQFH IURP RULJLQ LQ GHIRUPHG LPDJH &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < 127( 8QLWV DUH SL[HOV 7KH RULJLQ LV WKH XSSHU OHIW FRUQHU RI WKH LPDJH 6HH )LJXUH

PAGE 67

7DEOH 6SRW GLVWDQFH IURP LFRQ LQ XQGHIRUPHG LPDJH &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < 127( 8QLWV DUH SL[HOV 6SDFLQJ LV SL[HOV LQ [ DQG \ 6HH )LJXUH

PAGE 68

7DEOH 6SRW GLVWDQFH IURP LFRQ LQ GHIRUPHG LPDJH &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < r &2/ 52: ; < &2/ 52: ; < 127( 8QLWV DUH SL[HOV 6HH )LJXUH

PAGE 69

FKHFNV VDPH QXPEHU RI URZV LQ HDFK LPDJH VDPH QXPEHU RI FROXPQV LQ HDFK LPDJH VDPH QXPEHU RI VSRWV LQ HDFK LPDJH URZV WLPHV FROXPQV HTXDO QXPEHU RI VSRWV 2I FRXUVH LI DQ\ FKHFN IDLOV WKH LPDJHV DUH RXW RI FRQn WH[W ,I VWUDLQV DUH ODUJH HQRXJK FRQWH[W VLPLODULWLHV EHWZHHQ LPDJHV FDQQRW EH UHFRJQL]HG DQG DQ HUURU PHVVDJH LV SULQWHG $VVXPn LQJ ODUJH ORFDO URWDWLRQ GRHV QRW RFFXU LQ WKH PDWHULDO UHFRJQL]LQJ VLPLODU FRQWH[W EHWZHHQ WKH WZR SDWWHUQV WKDW GRHV QRW H[LVW LV YHU\ UHPRWH 7KXV LI QR FKHFN IDLOV WKH DSSUR[LPDWLRQ DOJRULWKP HVWDEn OLVKHV D RQHWRRQH UHODWLRQVKLS RU FRQWH[W PDSSLQJ EHWZHHQ VSRWV LQ WKH WZR LPDJHV

PAGE 70

&+$37(5 ,,, ,0$*( *(1(5$7,21 $1' $1$/<6,6 ,,, 6\QWKHWLF ,PDJHV &RPSXWHU JHQHUDWHG LPDJHV RU V\QWKHWLF LPDJHV ZHUH XVHG DV WKH SULPDU\ GHYHORSPHQW WRRO DQG WR VWXG\ WKH LQIOXHQFH RI YDULRXV SDUDPHWHUV 7KH RSWLRQV GDWD ILOH VKRZQ LQ 7DEOH ZDV XVHG WR JHQn HUDWH WKHVH ; SL[HO LPDJHV 0RWLRQ LV OLPLWHG WR DIILQH WUDQVIRUPDWLRQV SOXV ULJLGERG\ URWDWLRQ DERXW WKH LPDJH FHQWHU 7KH WUDQVIRUPDWLRQV DUH [ F; Ff< Ff f \ Ff; F< F 7KHUH DUH WKUHH LPDJH RXWSXW RSWLRQV DYDLODEOH 7KH %,1$5< RSWLRQ LV WKH PDFKLQH UHSUHVHQWDWLRQ RI WKH LPDJH '(&,0$/ LV WKH PDFKLQH YHUVLRQ FRQYHUWHG WR D UHDGDEOH IRUP VHH IRU H[DPSOH )LJXUH Ef 3/27 GLVSOD\V D WKUHVKROGHG LPDJH VXLWDEOH IRU KDUG FRS\ RQ D JUDSKLFV WHUPLQDO VHH IRU H[DPSOH )LJXUH f 8'$7 1,0*'$7 ''$7 DQG ',0*'$7 DUH XVHUGHILQHG ILOH QDPHV 7KH LPDJH JHQHUDWRU DXWRPDWLFDOO\ UHDGV WKH LQSXW ILOH DQG JHQHUDWHV WKH VSHFLILHG RXWSXW WKHUHE\ DOORZLQJ WKH XVHU D FKRLFH RI LQWHUDFWLYH RU EDWFK SURFHVVLQJ

PAGE 71

7DEOH ,QSXW ILOH IRU V\QWKHWLF LPDJH JHQHUDWRU ,0$*( *(1(5$725 ,1387 '$7$ 6327 *(20(75< 3$77(51 6,=( 5$',86 ;63$&,1* <63$&,1* ;&2/6 <52:6 *5$< /(9(/6 %/$&. :+,7( 12,6( '$7$ ',675,%87,21 7<3( 1 67$1'$5' '(9,$7,21 1250$/ 81,)250 ,0$*( 027,21 027,21 7<3( f§32/<120,$/ 21/< f§32/<120,$/ $1' 527$7,21 f§527$7,21 21/< 32/<120,$/ ; < 75$16/$7,21 [ \ 527$7,21 '(*5((6f ,0$*( 287387 )250$7 81'()250(' <1 '()250(' <1 %,1$5< 8'$7 < '(&,0$/ 1,0*'$7 1 3/27 r6&5((1r 1 ''$7 < ',0*'$7 1 m6&5((1r 1

PAGE 72

,,, ,PDJH $QDO\VLV 6LPLODU WR LPDJH JHQHUDWLRQ LPDJH DQDO\VLV EHJLQV ZLWK DQ RSn WLRQV ILOH ZKLFK LV VKRZQ LQ 7DEOH ,I 86( 7+(6( ),/( 1$0(6" LV DQVZHUHG QR WKH XVHU LV DXWRPDWLFDOO\ SURPSWHG IRU WKH ILOH QDPH 6LQFH QR RWKHU LQIRUPDWLRQ LV UHTXLUHG E\ WKH SURJUDP WKH XVHU LV IUHH WR FKRRVH LQWHUDFWLYH RU EDWFK SURFHVVLQJ 7DEOH ,PDJH DQDO\VLV RSWLRQV ILOH rrrrrrrr ,0$*( $1$/<6,6 ,1387287387 237,216 rrrrrrrr ,1387 63(&,),&$7,216 ,0$*( ),/( 1$0( 86( 7+(6( ),/( 1$0(6" 81'()250(' '()250(' 8 '$7 ''$7 1 1 ,0$*( 287387 81'()250(' '()250(' 8125'(5(' 6327 '$7$ < ',67$1&( )520 25,*,1 < ',67$1&( )520 ,&21 < < < < ',63/$&(0(17 $1$/<6,6 287387 727$/ ',63/$&(0(17 < ',63/$&(0(17 *5$',(176 < /$*5$1*,$1 675$,1 < 7$
PAGE 73

$IWHU VHJPHQWDWLRQ IHDWXUH H[WUDFWLRQ DQG FRQWH[W UHFRJQLWLRQ DXWRPDWLF SURFHVVLQJ FRQWLQXHV ZLWK PRWLRQ DQDO\VLV 7KH PRWLRQ RI HDFK VSRW LV UHVROYHG DQG WKH GLVSODFHPHQW LV FRPSXWHG E\ HTXDWLRQ f VHH 7DEOH f 1H[W WKH GHIRUPDWLRQ DQG GLVSODFHPHQW JUDGLHQWV DW HDFK VSRW DUH FRPSXWHG E\ GLIIHUHQWLDWLQJ WKH IXQFWLRQV XVHG WR GHVFULEH PRWLRQ DQG GLVSODFHPHQW 7DEOH 'LVSODFHPHQW RI HDFK VSRW &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < &2/ 52: ; < 127( 8QLWV DUH SL[HOV &RPSDUH LPDJHV LQ )LJXUH 'LVSODFHPHQWV DUH FRPSXWHG XVLQJ WKH GDWD LQ 7DEOHV DQG LQ HTXDWLRQ f 7KH FKRLFH RI D PDSSLQJ PRGHO GHSHQGV RI FRXUVH RQ WKH DSSOLFDWLRQ ,QWXLWLRQ D SULRUL NQRZOHGJH HODVWLFLW\ WKHRU\

PAGE 74

QXPHULFDO HIILFLHQF\ HWF DUH XVHG WR VHOHFW WKH DSSUR[LPDWLQJ IXQFn WLRQV ,Q WKLV VWXG\ VLPSOH SRO\QRPLDOV OLQHDU RU VHFRQG RUGHU DGHTXDWHO\ DSSUR[LPDWH WKH PRWLRQ DQG GLVSODFHPHQW LQ WKH V\QWKHWLF DQG UHDO LPDJHV 7KH PRGHOV DUH OLQHDU [ *; F 9 \ r; F 9 X F ; F F < Y r; r 9 f DQG VHFRQG RUGHU [ &; &; n &\ F < F;< \ F &; n &\ r< r;< X r[ r r; r r r : r r< r r;, 9 nXr r r; & &\ &< r;< f 7KHVH PRGHOV DUH ILWWHG WR D VPDOO QHLJKERUKRRG RI VSRWV DQG GLIIHUHQWLDWHG WR REWDLQ WKH GLVSODFHPHQW DQG GHIRUPDWLRQ

PAGE 75

JUDGLHQWV 7\SLFDOO\ WKH QHLJKERUKRRGV DUH ; ; RU ; DQG WKH JUDGLHQWV DUH FRPSXWHG IRU WKH FHQWHU VSRW ([FHSWLRQV DUH WKH VSRWV QHDU WKH HGJH RI WKH SDWWHUQ DQG WKHLU JUDGLHQWV DUH FRPSXWHG OLNH D IRUZDUGEDFNZDUG GLIIHUHQFH )RU ERWK PRGHOV WKH FRHIILFLHQWV DUH GHWHUPLQHG E\ OHDVW VTXDUHV HYDOXDWLRQ %DVHG RQ D PRGLILHG IRUP RI WKH &RQWH DQG 'HERRU >@ QRWDWLRQ WKH QRUPDO HTXDWLRQV DUH  f ZKHUH /LQHDU f ;
PAGE 76

F n /LQHDU 6HFRQG RUGHU f I \ Y X Y 1 1 1 f DQG 1 LV WKH QXPEHU RI VSRWV LQ WKH QHLJKERUKRRG /HW $ DQG I = (TXDWLRQV f EHFRPH $F = f ZKLFK DUH VROYDEOH E\ *DXVV-RUGDQ HOLPLQDWLRQ

PAGE 77

+RUQEHFN >@ SRLQWV RXW *DXVV-RUGDQ HOLPLQDWLRQ PD\ EH XVHG WR VROYH HTXDWLRQV f EXW XQIRUWXQDWHO\ WKH VHW LV H[WUHPHO\ LOOFRQGLWLRQHG 7DNLQJ KLV DGYLFH GRXEOH SUHFLVLRQ DULWKPHWLF LV XVHG WR FRSH ZLWK WKH ODUJH YDULDWLRQ LQ WKH PDJQLWXGHV RI WKH FRn HIILFLHQWV LQ DQ\ JLYHQ URZ ,Q DGGLWLRQ WR KLV VXJJHVWLRQV S DQG I DUH VFDOHG EHIRUH VROXWLRQ DQG WKH FnV DUH UHFRQYHUWHG GXULQJ JUDGLHQW FRPSXWDWLRQ )RU D YDULHW\ RI SURJUDPPLQJ UHDVRQV VROXWLRQ RI HTXDWLRQV f LV QRW SHUIRUPHG GLUHFWO\ E\ *DXVV-RUGDQ HOLPLQDWLRQ ,QVWHDG WKH LQYHUVH RI $ LV FRPSXWHG XVLQJ +RUQEHFNnV >@ *DXVV-RUGDQ PDWUL[ LQYHUVLRQ VXEURXWLQH ZLWK WKH DGGLWLRQ RI DSSURSULDWH HUURU WUDSV DQG F LV VROYHG E\ F $B= f +RZ DFFXUDWH DUH WKH YDOXHV RI F DQG KRZ ZHOO GR HTXDWLRQV f RU f ILW WKH GDWD" 'HEXJJLQJ H[SHULHQFH LQGLFDWHV WKH QXPEHU RI UHOLDEOH GLJLWV QXPHULFDOO\ VSHDNLQJ QRW H[SHULPHQWDOO\f UXQV LQ WKH WHHQV ,Q WKLV VWXG\ WKH VROXWLRQV DUH NQRZQ EXW LI WKH\ ZHUH XQNQRZQ JRRGQHVVRIILW DQG WKH FRQWULEXWLRQ RI HDFK WHUP FRXOG EH HDVLO\ WHVWHG E\ WKH PHWKRGV VXJJHVWHG E\ 0LOOHU DQG )UHXQG >@ 7KH H[DPSOH RI )LJXUH LV FRQWLQXHG ZLWK WKH FDOFXODWLRQ RI F $VVXPH WKH OLQHDU PRGHO RI HTXDWLRQ f DQG D ; QHLJKERUKRRG FHQWHUHG DW VSRW f $ VXEVWLWXWLRQ RI GDWD IURP 7DEOH LQWR HTXDWLRQ f SURGXFHV

PAGE 78

/LNHZLVH GDWD IURP 7DEOHV DQG LQ HTXDWLRQ f OHDG WR I 7KH VROXWLRQ IRU LV F f f 7KH PRWLRQ IXQFWLRQV LH WKH PDWKHPDWLFDO PDSVf LQ HTXDn WLRQV f EHFRPH [ ; < f \ ; <

PAGE 79

DQG WKH GLVSODFHPHQW PDSV LQ HTXDWLRQV f EHFRPH X ; < Y ; < f 7KH GHIRUPDWLRQ DQG GLVSODFHPHQW JUDGLHQWV FRPSXWHG E\ HTXDn WLRQV f DQG f UHVSHFWLYHO\ EHFRPH ) 6WUDLQ FRPSXWHG E\ HTXDWLRQV f RU f LV ( f f f %\ WKH 7D\ORU VHULHV PHWKRG RI HTXDWLRQ f VWUDLQ LV ( f

PAGE 80

6WUDLQ FRPSXWHG DFFRUGLQJ WR VPDOO GLVSODFHPHQW WKHRU\ LV LQ WKLV FDVH TXLWH LQDFFXUDWH %\ HTXDWLRQV f f 5LJLGERG\ URWDWLRQ LV FRPSXWHG DV IROORZV %\ HTXDWLRQ f WKH *UHHQ GHIRUPDWLRQ WHQVRU LV & 7KH HLJHQYDOXHV RI & DUH & 3 DQG & LV 3 6LQFH 8 & WKHQ 3 a3 8 f f f f

PAGE 81

5RWDWLRQ WHQVRU 5 FRPSXWHG E\ HTXDWLRQ f LV 5 7KH UHPDLQGHU RI WKH RXWSXW IRU WKH H[DPSOH LQ )LJXUH f LV VKRZQ LQ 7DEOHV WKURXJK

PAGE 82

7DEOH 'LVSODFHPHQW JUDGLHQWV &2/ 52: X; Y; X< Y
PAGE 83

7DEOH /DJUDQJLDQ VWUDLQ IURP GLVSODFHPHQW JUDGLHQWV &2/ 52: A ( ( &2/ 52: ( ( ( &2/ 52: ( ( S ( &2/ 52: ( ( n ( &2/ 52: ( (" ( 127( 8QLWV DUH PLFUR LQFKHV SHULQFK &RUUHFW YDOXHV DUH (8

PAGE 84

7DEOH 6WUDLQ IURP 7D\ORU VHULHV &2/ 52: A ec ( &2/ 52: ( ( ( &2/ 52: ( ( ( &2/ 52: ( ( ( f &2/ 52: ( ( ( 127( 8QLWV DUH PLFUR LQFKHV ( ( SHULQFK &RUUHFW YDOXHV DUH ( (

PAGE 85

7DEOH 'HIRUPDWLRQ JUDGLHQW &2/ 52: ) ) ) ) &2/ 52: ) ) ) ) &2/ 52: ) ) ) ) &2/ 52: 0 ) ) &2/ 52: ) ) I 127( &RUUHFW YDOXHV DUH ) )Q ) )

PAGE 86

7DEOH *UHHQ GHIRUPDWLRQ WHQVRU &2/ 52: & F & &2/ 52: & & & &2/ 52: & & & &2/ 52: & & & &2/ 52: & & & 127( &RUUHFW YDOXHV DUH & &

PAGE 87

7DEOH 5LJKW VWUHWFK WHQVRU &2/ 52: 8 X X &2/ 52: 8 8 X &2/ 52: 8 8 8 &2/ 52: 8 8 8 &2/ 52: 8 8 8 127( &RUUHFW YDOXHV DUH ,/ A 8SS r 8WS

PAGE 88

7DEOH 5RWDWLRQ WHQVRU &2/ 52: 5 5 5 5 &2/ 52: 5 5 5 5 a &2/ 52: 5 5 5 5 &2/ 52: 5 5 5 5 &2/ 52: 5 5 5 5 127( &RUUHFW YDOXHV DUH 5 arf 5 5 5

PAGE 89

7DEOH /DJUDQJLDQ VWUDLQ IURP GHIRUPDWLRQ JUDGLHQW &2/ 52: ( ( S ( &2/ 52: ( ( ( &2/ 52: ( (" ( &2/ 52: ( ( S ( &2/ 52: ( ( S (  127( 8QLWV DUH PLFUR LQFKHV SHULQFK &RUUHFW YDOXHV DUH (LM

PAGE 90

&+$37(5 ,9 (;3(5,0(176 ,9 7HVW 6SHFLPHQ 7KH SULVPDWLF EHDP VKRZQ LQ )LJXUH ZDV XVHG IRU DOO H[SHULn PHQWV $ FOHDU SODVWLF &5 ZDV FKRVHQ IRU LWV UHODWLYHO\ ORZ PRGXOXV RI HODVWLFLW\ DQG FDSDFLW\ IRU ODUJH HODVWLF VWUDLQ 7KH ; WHVW SDWWHUQ RQ WKH EHDP ZDV JHQHUDWHG E\ WKH V\QWKHn WLF LPDJH JHQHUDWRU DQG DSSOLHG E\ FRQWDFW SULQWLQJ )LUVW D KDUG FRS\ RI WKH SDWWHUQ ZDV SKRWRJUDSKHG XVLQJ .RGDOLWK ILOP $IWHU FRDWLQJ WKH EHDP ZLWK D EXIIHU RI SRO\XUHWKDQH DQG QDSKWKD D /LTXLG /LJKW HPXOVLRQ ZDV DSSOLHG FRQWDFW SULQWHG DQG FKHPLFDOO\ IL[HG VHH )LJXUH f ,9 7HVW (TXLSPHQW 7KUHH H[SHULPHQWDO VHWXSV ZHUH FRQILJXUHG WZR ZHUH IRU VWUDLQ WHVWV DQG WKH WKLUG IRU ULJLGERG\ PRWLRQ 7KH VWUDLQ VHWXS ZDV D W\SLFDO SKRWRHODVWLF ORDG IUDPH ZLWK VSHFLDO IL[WXUHV VHH )LJXUH f 7KH IL[WXUHV ZHUH IRU IRXUSRLQW EHQGLQJ VHH )LJXUH f DQG FDQWLOHYHU EHDP VHH )LJXUH f H[SHULn PHQWV 7KH HTXLSPHQW VKRZQ LQ )LJXUH ZDV XVHG IRU WKH ULJLGERG\ PRWLRQ WHVWV $ SKRWRFRS\ RI D SURWUDFWRU ZDV WDSHG WR WKH EDFN RI WKH OHQV KROGHU DQG URWDWLRQ ZDV PHDVXUHG UHODWLYH WR WKH WLS RI WKH

PAGE 91

79 / )LJXUH %HDP IDEULFDWHG IURP &5 )LJXUH 6SRW SDWWHUQ RQ EHDP

PAGE 92

( f Q D ‘ I O/ )LJXUH ([SHULPHQWDO VHWXS RI ORDG IUDPH )LJXUH 6HWXS IRU IRXUSRLQW EHQGLQJ

PAGE 93

)LJXUH 6HWXS IRU FDQWLOHYHU EHDP )LJXUH 5LJLGERG\ PRWLRQ VHWXS

PAGE 94

FODPSHG SRLQWHU 0HDVXUHPHQWV ZHUH DFFXUDWH WR DERXW GHJUHH $ GLDO LQGLFDWRU PHDVXUHG WUDQVODWLRQ LQ WKH KRUL]RQWDO GLUHFWLRQ ZLWK DQ DFFXUDF\ RI s LQFK 6LQFH LPDJH SURFHVVLQJ HTXLSPHQW ZDV XQDYDLODEOH DW WKH 8QLYHUVLW\ RI )ORULGD WKH HTXLSPHQW IRU WKH ULJLGERG\ H[SHULPHQWV ZDV WUDQVSRUWHG WR 86& ZKHUH WKH GLJLWDO LPDJHV ZHUH UHFRUGHG ,9 ([SHULPHQWDO 3URFHGXUH 8QIRUWXQDWHO\ WKH VWUDLQ WHVWV ZHUH QRW GRQH EHFDXVH WKH ORDG IUDPH ZDV WRR ODUJH WR WUDQVSRUW WR 86& 7KH VHFRQG RUGHU PRGHO LQ HTXDWLRQ f ZDV IRU WKHVH WHVWV 7KH HTXLSPHQW VKRZQ LQ )LJXUH DQG DQ LPDJH SURFHVVLQJ V\VWHP VLPLODU WR WKH RQH VKRZQ LQ )LJXUH EDVHG RQ WKH ,%0 3&f ZDV XVHG IRU WKH ULJLGERG\ H[SHULPHQWV 7KH &&' FDPHUD ZDV D 6RQ\ PRGHO ;&0 ZLWK D ]RRP OHQV DQG WKH GLJLWL]HU ERDUG ZDV DQ HLJKWELW 'DWDFXEH )UDPH *UDEEHU ZLWK URZV DQG FROXPQV RI SL[HOV 7KH VHWXS LQFOXGHG VHYHUDO URXWLQH FKHFNV )LUVW DOLJQPHQW EHWZHHQ WKH UDLO WDEOH DQG FDPHUD ZHUH FKHFNHG WR PLQLPL]H H[SHULn PHQWDO HUURU 7KHQ WKH ]RRP OHQV ZDV DGMXVWHG WR HQODUJH WKH SDWWHUQ LPDJH DV PXFK DV SRVVLEOH 5HVROXWLRQ LQ WKH KRUL]RQWDO GLUHFWLRQ ZDV PHDVXUHG DW SL[HOV SHU LQFK /LJKWLQJ ZDV DGMXVWHG WR SURYLGH WKH XVXDO LOOXPLQDWLRQ IRU WHVWV LQ WKH 86& ODE DQG WKH WUDQVSDUHQW VSHFLPHQ FDXVHG VRPH OLJKWLQJ SUREOHPV &RQWUDVW ZDV LPSDLUHG E\ WKH OLJKW WUDQVPLWWHG WKURXJK WKH EHDP

PAGE 95

$IWHU VHWXS LPDJHV ZHUH GLJLWL]HG DQG VWRUHG RQ IORSS\ GLVN $OO LPDJHV ZHUH WUDQVIHUUHG WR PDJQHWLF WDSH YLD 86&nV 9$; FRPSXWHU IRU WUDQVSRUW WR WKH 8QLYHUVLW\ RI )ORULGD

PAGE 96

&+$37(5 9 $1$/<6,6 2) 5(68/76 7KH DQDO\VLV RI UHVXOWV KDV EHHQ GLYLGHG LQWR WZR VHFWLRQV QXPHULFDO DQG H[SHULPHQWDO 7KH QXPHULFDO VHFWLRQ VXPPDUL]HV V\QWKHn WLF LPDJH ZRUN DQG WKH H[SHULPHQWDO VHFWLRQ VXPPDUL]HV WKH DQDO\VLV RI WKH LPDJHV GLJLWL]HG DW 86& 9 1XPHULFDO 7KH DQDO\VLV RI V\QWKHWLF LPDJHV LQFOXGHG ULJLGERG\ PRWLRQ KRPRJHQHRXV VWUDLQ FRPELQHG PRWLRQ DQG WKH LQIOXHQFH RI QRLVH JUD\ OHYHOV DQG VSRW UDGLXV 7R HQVXUH D JRRG UDQGRP VDPSOH WKH UHVXOWV IURP D YDULHW\ RI VSRW VSDFLQJV ZHUH DYHUDJHG 7KH SDWWHUQ ZDV ; ZLWK WKH LFRQ EHLQJ VSRW f ([FHSW DV QRWHG WKH JUD\ OHYHO GLIIHUHQFH ZDV DQG WKH ]HURPHDQ QRLVH ZDV *DXVVLDQ ZLWK D VWDQGDUG GHYLDWLRQ RI JUD\ OHYHOV 7KH DEVROXWH YDOXHV RI QRLV\ SL[HOV ZLWK YDOXHV OHVV WKDQ ]HUR ZHUH XVHG LQVWHDG RI WKH QHJDWLYH JUD\ OHYHOV ,Q WKH IROORZLQJ WDEOHV WKH DYHUDJH DEVROXWH HUURU LV GHILQHG DV IROORZV DYHUDJH DEVROXWH HUURU LV WKH DYHUDJH RI WKH DEVROXWH YDOXHV RI WKH HUURUV )RU H[DPSOH LI DQG WKH FRPSXWHG YDOXHV DUH (A \ (A af\ DQG ( S WKH DYHUDJH DEVROXWH HUURU LV \

PAGE 97

6WDQGDUG GHYLDWLRQ DOVR LQ WKH WDEOHV LV WKH VWDQGDUG GHYLDn WLRQ RI WKH FRPSXWHG YDOXHV )RU WKH VWUDLQ H[DPSOH DERYH WKH VWDQGDUG GHYLDWLRQ LV \ )RU WUDQVODWLRQV RI YDULRXV DPRXQWV LQ WKH [ DQG \ GLUHFWLRQV WKH DYHUDJH DEVROXWH HUURU ZDV SL[HOV ZLWK D VWDQGDUG GHYLDWLRQ RI SL[HOV 7UDQVODWLRQV QHDU ZKROH SL[HOV KDG VOLJKWO\ OHVV HUURU (TXDWLRQ f VXJJHVWHG D VSDWLDO VHQVLWLYLW\f 5LJLGERG\ URWDWLRQ ZDV DERXW WKH LPDJH FHQWHU ZKLFK ZDV DSSUR[n LPDWHO\ WZRWKLUGV RI D SL[HO EHORZ WKH SDWWHUQ FHQWHU 5RWDWLRQV UDQJHG IURP WR s GHJUHHV DQG WKH UHVXOWV FRPSXWHG XVLQJ WKH HOHPHQWV RI WKH URWDWLRQ WHQVRU ZHUH YHU\ FRQVLVWHQW ZLWK DQ DYHUDJH DEVROXWH HUURU RI GHJUHHV DQG D VWDQGDUG GHYLDWLRQ RI GHJUHHV 6HYHUDO VWUDLQ FDVHV ZHUH LQYHVWLJDWHG DQG VXPPDUL]HG LQ 7DEOH 1RUPDO VWUDLQV ZHUH A ( ZLWO ( ( A VKHDU VWUDLQV ZHUH ( ( ZAK ( H DQG FRPELQHG VWUDLQV ZHUH (_ M ( ( ( f 6HH HTXDWLRQ f IRU VWUDLQ GHILQLWLRQV 1RLVH KDG D VLJQLILFDQW HIIHFW RQ UHVXOWV $V VXJJHVWHG E\ HTXDWLRQ f LQFUHDVLQJ 615 GHFUHDVHG WUDQVODWLRQ DQG VWUDLQ HUURU VHH 7DEOH f )UDPLQJ IDLOXUHV EHJDQ RFFXUULQJ EHORZ 615 *UD\ OHYHO ZDV DQRWKHU VLJQLILFDQW YDULDEOH EHFDXVH RI WKH 615 DQG WKH GLJLWL]DWLRQ RI LOOXPLQDWLRQ LQWHQVLW\ )RU FRQVWDQW QRLVH LQFUHDVLQJ JUD\ OHYHO GLIIHUHQFH LQFUHDVHG WKH 615 DQG WKHUHE\ LQn FUHDVHG DFFXUDF\ :LWK D QRLVH VWDQGDUG GHYLDWLRQ RI JUD\ OHYHOV VSRWV ZHUH IUHTXHQWO\ XQLGHQWLILDEOH EHORZ D JUD\ OHYHO

PAGE 98

7DEOH 6XPPDU\ RI VWUDLQ DQDO\VLV IRU V\QWKHWLF LPDJHV 6WUDLQ &DVH 'LVSODFHPHQW *UDGLHQW DXLD[M 6WUDLQ &RUUHFW $YH $EV (UURU 6WDQGDUG 'HYLDWLRQ 1RUPDO 6KHDU D &RPELQHG D /LPLWHG E\ SDWWHUQ JHRPHWU\ 127( 5LJLGERG\ PRWLRQ ZDV ]HUR 6WUDLQ XQLWV DUH PLFUR LQFKHV SHULQFK 615

PAGE 99

7DEOH 7KH HIIHFW RI 615 RQ WUDQVODWLRQ DQG VWUDLQ FDOFXODWLRQV 7UDQVODWLRQ 6WUDLQ $YH $EV 6WDQGDUG $YH $EV 6WDQGDUG 615 (UURU 'HYLDWLRQ (UURU 'HYLDWLRQ 127( 7UDQVODWLRQV ZHUH SL[HOV LQ [ DQG \ 1R URWDWLRQ $OO VWUDLQV ZHUH (sM \ 7DEOH 7KH HIIHFW RI JUD\ OHYHO GLIIHUHQFH RQ WUDQVODWLRQ DQG VWUDLQ FDOFXODWLRQV 7UDQVODWLRQ 6WUDLQ $YH $EV 6WDQGDUG $YH $EV 6WDQGDUG + (UURU 'HYLDWLRQ (UURU 'HYLDWLRQ rr 127( 7UDQVODWLRQV ZHUH SL[HOV LQ [ DQG \ 1R URWDWLRQ $OO VWUDLQV ZHUH (LM LS

PAGE 100

GLIIHUHQFH RI %HORZ D GLIIHUHQFH RI D VWUDLQ RI (MM \ ZDV XQGHWHFWDEOH VHH 7DEOH f $Q LQFUHDVH LQ JUD\ OHYHO GLIIHUHQFH UHGXFHG WKH HUURU FDXVHG E\ GLJLWL]DWLRQ :LWKRXW QRLVH D GLIIHUHQFH RI JUD\ OHYHOV SURGXFHG DQ DYHUDJH DEVROXWH VWUDLQ HUURU RI \ IRU (LM \f ZLWK D VWDQGn DUG GHYLDWLRQ LM DQG DERYH D GLIIHUHQFH RI HUURU ZDV QLO +RZHYHU 615 HIIHFWV JUHDWO\ RYHUVKDGRZHG GLJLWL]DWLRQ HUURU 7KH HIIHFW RI VSRW UDGLXV GHFUHDVHG UDSLGO\ ZLWK LQFUHDVHG UDGLXV VHH 7DEOH f (TXDWLRQ f VXJJHVWHG WKLV IRU VPDOO ( DQG LQ IDFW DOPRVW QR FKDQJH RFFXUUHG IRU UDGLL DERYH SL[HOV 7DEOH 7KH HIIHFW RI VSRW UDGLXV RQ WUDQVODWLRQ DQG VWUDLQ FDOFXODWLRQV 7UDQVODWLRQ 6WUDLQ 6SRW $YH $EV 6WDQGDUG $YH $EV 6WDQGDUG 5DGLXV (UURU 'HYLDWLRQ (UURU 'HYLDWLRQ 127( VWUDLQV 7UDQVODWLRQV ZHUH ZHUH (LM \ 615 SL[HOV LQ [ DQG \ 1R URWDWLRQ $OO +RZ PXFK 9$; FRPSXWHU WLPH GLG WKH DQDO\VLV RI D ; SL[HO LPDJH UHTXLUH" )RU D ; VSRW SDWWHUQ &38 WLPH ZDV DERXW

PAGE 101

VHFRQGV ZLWK QR SULQWRXW ZULWWHQ WR GLVN DQG DERXW VHFRQGV ZLWK D IXOO SULQWRXW VHH 7DEOHV DQG f $Q ; SDWWHUQ UHTXLUHG DERXW VHFRQGV IRU QR SULQWRXW DQG DERXW VHFRQGV IRU D IXOO SULQWRXW 'XH WR RYHUDOO V\VWHP ORDG &38 WLPHV YDULHG DV PXFK DV SHUFHQW 9 ([SHULPHQWDO ,PDJHV GLJLWL]HG DW 86& ZHUH IRU ULJLGERG\ H[SHULPHQWV 7KHVH LQFOXGHG URWDWLRQV RI DQG GHJUHHV KRUL]RQWDO \ GLUHFWLRQf WUDQVODWLRQV RI DQG SL[HOV DQG D FRPn ELQHG PRWLRQ RI GHJUHHV URWDWLRQ SOXV SL[HOV WUDQVODWLRQ LQ WKH KRUL]RQWDO GLUHFWLRQ 7KH LPDJHV GLJLWL]HG GXULQJ WKH H[SHULn PHQWV ZHUH VXPPDUL]HG LQ 7DEOH 7DEOH ,PDJHV XVHG IRU ULJLGERG\ PRWLRQ H[SHULPHQWV ,PDJH *UD\ %ODFN /HYHO :KLWH DQ + 615 8QGHIRUPHG 8 8 8 'HIRUPHG ' ' ' 127( *UD\ OHYHOV D DQG + Q ZHUH HVWLPDWHG E\ WKH DQDO\VLV SURJUDP

PAGE 102

,Q VSLWH RI UHODWLYHO\ ORZ 615 WKH LPDJHV ZHUH VWLOO XVDEOH 7KH 615 IRU WKH LPDJHV ZHUH QHDU WKH PLQLPXP XVDEOH OHYHO VXJJHVWHG E\ V\QWKHWLF LPDJH DQDO\VLV ,Q IDFW XQGHIRUPHG LPDJHV 8 DQG 8 UHTXLUHG D VLPSOH DG KRF DGMXVWPHQW WR RQH RI WKHLU PDSSLQJ SDUDPHWHUV 2WKHUZLVH DQDO\VLV SURFHHGHG DV H[SHFWHG 5RWDWLRQV DFFRPSOLVKHG XVLQJ WKH OHQV KROGHU SUHYLRXVO\ GHn VFULEHG ZHUH DERXW D SRLQW QHDU WKH LPDJH FHQWHU $V VKRZQ LQ 7DEOH WKH DYHUDJH UHVXOWV H[FHSW IRU WKH WKLUG WHVW ZHUH ZLWKLQ WKH GHJUHH DFFXUDF\ RI WKH H[SHULPHQW 7KH DYHUDJH DQG WKH DYHUDJH DEVROXWH VWUDLQ HUURU ZDV FRPSXWHG XVLQJ DOO WKH VWUDLQV LH (_M (! DQG ( DW D V3rWVf FRPSXWHG E\ WKH DQDO\VLV SURJUDP 7KH DFWXDO VWUDLQ IRU DQ\ 7DEOH 5HVXOWV RI ULJLGERG\ URWDWLRQ H[SHULPHQWV ,PDJH 5RWDWLRQ 6WUDLQ (UURU $FWXDO $YH (UURU 6WG 'HY $YH $YH $EV ' ' D D GHJUHHV SOXV SL[HOV KRUL]RQWDO WUDQVODWLRQ 127( 7KH XQGHIRUPHG LPDJH ZDV 8 IRU WKH ILUVW WKUHH H[SHULPHQWV DQG 8 IRU WKH ODVW $OO URWDWLRQV ZHUH LQ GHJUHHV 6WUDLQ XQLWV DUH PLFUR LQFKHVSHULQFK 0HDVXUHPHQW DFFXUDF\ ZDV DERXW GHJUHH

PAGE 103

ULJLGERG\ PRWLRQ ZDV RI FRXUVH $OWKRXJK WKH QRLVH ZDV *DXVVLDQ WKH DYHUDJH VWUDLQ HUURU ZDV QRW WKH H[SHFWHG YDOXH RI 7KH UHDVRQ IRU WKLV ZDV LPDJLQJ V\VWHP GLVWRUWLRQ $YHUDJH UHVXOWV REWDLQHG GXULQJ WUDQVODWLRQ H[SHULPHQWV DUH VXPPDUL]HG LQ 7DEOH 7KH ILUVW WHVW ZDV ZLWKLQ WKH s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n WRUWLRQ ZDV LQGXFHG E\ WKH LPDJH SURFHVVLQJ V\VWHP 2WKHU VRXUFHV RI HUURU ZHUH ORZ 615 ORZ JUD\ OHYHO GLIIHUHQFH DQG UHODWLYHO\ FRDUVH UHVROXWLRQ RI WKH H[SHULPHQW VHWXS $OWKRXJK D FRPPRQ VROXWLRQ IRU PLQLPL]LQJ QRLVH HIIHFWV LV DYHUDJLQJ PXOWLSOH LPDJHV RQO\ RQH FRS\ RI HDFK LPDJH ZDV XVHG LQ WKHVH H[SHULPHQWV $ VPDOO LQFUHDVH LQ JUD\ OHYHO GLIIHUHQFH ZDV SRVVLEOH WKURXJK WKH XVH RI DGGLWLRQDO OLJKWLQJ RU E\ XVLQJ DQ RSDTXH UHIOHFWLYH

PAGE 104

7DEOH 5HVXOWV RI ULJLGERG\ WUDQVODWLRQ H[SHULPHQWV ,PDJH 7UDQVODWLRQ 6WUDLQ (UURU +RUL]RQWDO 9HUWLFDO $FWXDO $YH (UURU 6WG 'HY $FWXDO $YH (UURU 6WG 'HY $YH $YH $EV ' 127( 7KH XQGHIRUPHG LPDJH ZDV 8 DQG DOO WUDQVODWLRQV DUH LQ SL[HOV 6WUDLQ XQLWV DUH PLFUR LQFKHVSHULQFK 0HDVXUHPHQW DFFXUDF\ ZDV s SL[HOV U?M!

PAGE 105

VSHFLPHQf ,PSURYHG UHVROXWLRQ ZDV QRW SRVVLEOH GXH WR WKH OLPLWDn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f ZHUH SRVVLEOH LI WKH XQGHIRUPHG LPDJH ZHUH DQDO\]HG RQO\ RQFH

PAGE 106

&+$37(5 9, &21&/86,216 $1' 5(&200(1'$7,216 9, &RQFOXVLRQV 7KH QXPHULFDO DQG H[SHULPHQWDO UHVXOWV VXFFHVVIXOO\ GHPRQVWUDWHG WKH SRWHQWLDO RI SDWWHUQ PDSSLQJ LQ PRWLRQ DQDO\VLV :LWK WKLV PHWKRG LW ZDV SRVVLEOH WR REWDLQ DFFXUDWH PHDVXUHPHQWV RI URWDWLRQ WUDQVODWLRQ DQG VWUDLQ IRU D YDULHW\ RI SODQH PRWLRQV VHH 7DEOHV f $XWRPDWLRQ PLQLPL]HG XVHU WLPH DQG UHTXLUHG RQO\ D PRGHVW DPRXQW RI FRPSXWHU UHVRXUFHV :KROH LPDJH DQDO\VLV ZDV DOVR WKRURXJKO\ GHPRQVWUDWHG 6HJPHQWDWLRQ DOJRULWKPV SHUIRUPHG ZHOO LQ DOO LPDJHV WHVWHG 7KH HVWLPDWLRQ RI KLVWRJUDP SDUDPHWHUV ZDV KLJKO\ HIILFLHQW DQG SURn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

PAGE 107

H[SHULPHQWDO UHVXOWV VXJJHVWHG D SUDFWLFDO PLQLPXP 615 EHWZHHQ DQG $OWKRXJK D ODFN RI GDWD SUHYHQWV D GLUHFW FRPSDULVRQ RI QXPHULFDO DQG H[SHULPHQWDO UHVXOWV WKH\ DJUHHG GLUHFWLRQDOO\ IRU URWDWLRQ DQG WUDQVODWLRQ ,Q JHQHUDO VSRW UDGLXV ZDV QRW DQ LPSRUWDQW YDULDEOH 3UDFWLn FDO H[SHULHQFH DQG QXPHULFDO UHVXOWV LQGLFDWHG UDGLL VKRXOG EH DW OHDVW WR SL[HOV 3DWWHUQ PDSSLQJ KDV WKH SRWHQWLDO IRU VLJQLILFDQW GDWD FRPSUHVn VLRQ 2WKHU LPDJH SURFHVVLQJ DSSOLFDWLRQV UHTXLUH DUFKLYLQJ HQWLUH LPDJHV ,I WKH SDWWHUQ LV ZHOO VXLWHG WR WKH DSSOLFDWLRQ WKH HQWLUH GLVSODFHPHQW ILHOG PD\ EH DGHTXDWHO\ UHFRQVWUXFWHG IURP WKH VSRW LQIRUPDWLRQ 9, 5HFRPPHQGDWLRQV $OWKRXJK WKH H[SHULPHQWDO GDWD LQFOXGHG LQ WKLV VWXG\ LV HQFRXUn DJLQJ SDWWHUQ PDSSLQJ QHHGV H[WHQVLYH H[SHULPHQWDO YHULILFDWLRQ ZLWK D ZLGH YDULHW\ RI SODQH PRWLRQ DSSOLFDWLRQV 2QO\ E\ XVLQJ D ULJRUn RXVO\ FDOLEUDWHG V\VWHP DQG DQDO\]LQJ PDQ\ LPDJHV FDQ WKH IXOO SRWHQWLDO RI WKLV PHWKRG EH NQRZQ $ ORJLFDO QH[W VWHS LV WKH FRQVLGHUDWLRQ RI LPDJH SURFHVVLQJ DQG SDWWHUQ UHFRJQLWLRQ WRROV VXFK DV ILOWHULQJ ORFDO DYHUDJLQJ RWKHU GLPHQVLRQV RI IHDWXUH VSDFH HWF 7KHVH PD\ VLJQLILFDQWO\ LPSURYH SHUIRUPDQFH LQ ORZ 615 VLWXDWLRQV RU OHDG WR WKH GLVFRYHU\ RI QHZ SDUDPHWHUV DQG PRUH JHQHUDO PDSSLQJ WKHRULHV 7KH LPSOHPHQWDWLRQ RI WKH IXOO V\QWDFWLF UHFRJQLWLRQ WKHRU\ VKRXOG DOVR EH LQYHVWLJDWHG

PAGE 108

$Q H[WHQVLRQ RI WKLV PHWKRG WR WKUHHGLPHQVLRQDO PRWLRQ DQG PDSSLQJ RI PXOWLSOH LPDJHV PRVDLFVf LV DOVR VXJJHVWHG 7KUHH GLPHQVLRQDO LPDJHV WKDW DUH QRW UHVWULFWHG E\ FDPHUD YLHZ KDYH PXFK SRWHQWLDO IRU SUDFWLFDO DSSOLFDWLRQ $V D ILQDO QRWH LW LV VXJJHVWHG WKDW SDWWHUQ PDSSLQJ EH LQWHn JUDWHG ZLWK RWKHU H[SHULPHQWDO PHWKRGV DQG QXPHULFDO DQDO\VLV WRROV 7KLV LQFOXGHV DGYDQFHG JUDSKLFV IRU GDWD PDQLSXODWLRQ DQG SUHVHQWDWLRQ

PAGE 109

$33(1',; %25'(5 )2//2:,1* $/*25,7+0

PAGE 110

,67$57O ! .;
PAGE 111

5()(5(1&(6 'DOO\ : DQG 5LOH\ : ) ([SHULPHQWDO 6WUHVV $QDO\VLV 0F*UDZ+LOO 1HZ
PAGE 112

9RORVKLQ $ 6 DQG %XUJHU & 3 +DOIIULQJH 3KRWRHODVWLn FLW\ $ 1HZ $SSURDFK WR :KROH)LHOG 6WUHVV $QDO\VLV ([S 0HFK f :HDWKHUV 0 )RVWHU : $ 6ZLQVRQ : ) DQG 7XUQHU / ,QWHJUDWLRQ RI /DVHU6SHFNOH DQG )LQLWH(OHPHQW 7HFKQLTXHV RI 6WUHVV $QDO\VLV ([S 0HFK f *RQ]DOHV 5 & DQG :LQW] 3 'LJLWDO ,PDJH 3URFHVVLQJ $GGLVRQ:HVOH\ 5HDGLQJ 0$ f 0HQGHQKDOO ) 7 &RPSXWHU $LGHG $QDO\VLV RI 3KRWRHODVWLF ,PDJHV 3K' 'LVVHUWDWLRQ 8QLY RI ,OOLQRLV 8UEDQD f 3'3 3URFHVVRU +DQGERRN 'LJLWDO (TXLSPHQW &RUS 0D\QDUG 0$ f 5RVHQIHOG $ 'LJLWDO 3LFWXUH 3URFHVVLQJ YROV DQG $FDGHPLF 3UHVV 1HZ
PAGE 113

0LOOHU $ ,PSURYHG 3KRWRJULG 7HFKQLTXHV IRU 'HWHFWLRQ RI 6WUDLQ 2YHU 6KRUW *DJH /HQJWKV 3URF 6(6$ B f 2EDWD 0 6KLPDGD + DQG .DZDVDNL $ )LQH*ULG 0HWKRG IRU /DUJH 6WUDLQ $QDO\VLV 1HDU D 1RWFK 7LS ([S 0HFK f 2EDWD 0 DQG 6KLPDGD + (PEHGGHG )LQH*ULG 0HWKRG $SSOLHG WR 7KUHH'LPHQVLRQDO 6WUHVV,QWHQVLW\)DFWRU $QDO\VLV ([S 0HFK f 3DUNV 9 7KH *ULG 0HWKRG ([S 0HFK 11 f 3DUNV 9 6WUDLQ 0HDVXUHPHQW 8VLQJ *ULGV 2SW (QJ f 6HYHQKXLMHVHQ 3 :H 1HHG 6LPSOLFLW\ DQG 6SHHG 3UHVHQWHG DW %6605$6 -RLQW &RQI 1/5 $PVWHUGDP f 6HYHQKXLMHVHQ 3 7ZR 6LPSOH 0HWKRGV IRU 'HIRUPDWLRQ 'HPRQn VWUDWLRQ 0HDVXUHPHQW 6WUDLQ f %DOODUG $ DQG %URZQ & 0 &RPSXWHU 9LVLRQ 3UHQWLFH +DOO (QJOHZRRG &OLIIV 1f 0F9H\ ( 6 DQG /HH : 6RPH $FFXUDF\ DQG 5HVROXWLRQ $VSHFWV RI &RPSXWHU 9LVLRQ 'LVWDQFH 0HDVXUHPHQWV ,((( 7UDQV 3$0, 3$0, f 0XUL $ = 2Q 0RQRFXODU 3HUFHSWLRQ RI f' 0RYLQJ 2EMHFWV ,((( 7UDQV 3$0, 3$0, f 5RDFK : DQG $JJDUZDO 'HWHUPLQLQJ WKH 0RYHPHQW RI 2EMHFWV IURP D 6HTXHQFH RI ,PDJHV ,((( 7UDQV 3$0, 3$0, f +XDQJ 7 6 DQG 7VDL 5 < f' 0RWLRQ (VWLPDWLRQ IURP ,PDJH 6SDFH 6KLIWV ,((( 3URF ,&$663 f 7VDL 5 < DQG +XDQJ 7 6 (VWLPDWLQJ 7KUHH'LPHQVLRQDO 0RWLRQ 3DUDPHWHUV RI D 5LJLG 3ODQDU 3DWFK ,((( 7UDQV $663 $663 f 7VDL 5 < DQG +XDQJ 7 6 (VWLPDWLQJ 7KUHH'LPHQVLRQDO 0RWLRQ 3DUDPHWHUV RI D 5LJLG 3ODQDU 3DWFK ,, 6LQJXODU 9DOXH 'HFRPSRVLWLRQ ,((( 7UDQV $663 $663 f 7VDL 5 < DQG +XDQJ 7 6 (VWLPDWLQJ 7KUHH'LPHQVLRQDO 0RWLRQ 3DUDPHWHUV RI D 5LJLG 3ODQDU 3DWFK ,,, )LQLWH 3RLQW &RUUHVSRQGHQFHV DQG WKH 7KUHH9LHZ 3UREOHP ,((( 7UDQV $663 $663 f

PAGE 114

7VDL 5 < DQG +XDQJ 7 6 8QLTXHQHVV DQG (VWLPDWLRQ RI 7KUHH'LPHQVLRQDO 0RWLRQ 3DUDPHWHUV RI D 5LJLG 2EMHFWV ,((( 7UDQV 3$0, 3$0, r f (ULQJHQ $ & 0HFKDQLFV RI &RQWLQXD :LOH\ 1HZ
PAGE 115

• 2WVX 1 $ 7KUHVKROG 6HOHFWLRQ 0HWKRG )URP *UD\/HYHO +LVWRn JUDPV ,((( 7UDQV 60& 60& f :HV]ND 6 1DJHO 5 1 DQG 5RVHQIHOG $ $ 7KUHVKROG 6HOHFWLRQ 7HFKQLTXH ,((( 7UDQV &RPSXW & f :HV]ND 6 DQG 5RVHQIHOG $ 7KUHVKROG (YDOXDWLRQ 7HFKQLn TXHV ,((( 7UDQV 60& 60& f :HV]ND 6 DQG 5RVHQIHOG $ +LVWRJUDP 0RGLILFDWLRQ IRU 7KUHVKROG 6HOHFWLRQ ,((( 7UDQV 60& 60& f 0LOOHU DQG )UHXQG ( 3UREDELOLW\ DQG 6WDWLVWLFV IRU (QJLQHHUV 6HFRQG (G 3UHQWLFH+DOO (QJOHZRRG &OLIIV 1f )XNXQDJD ,QWURGXFWLRQ WR 6WDWLVWLFDO 3DWWHUQ 5HFRJQLWLRQ $FDGHPLF 3UHVV 1HZ
PAGE 116

%,2*5$3+,&$/ 6.(7&+ 5 :DOODFH )DLO ZDV ERUQ LQ 0RELOH $ODEDPD RQ $XJXVW +H JUHZ XS LQ VRXWKHUQ $ODEDPD DQG JUDGXDWHG IURP )RUW 'DOH $FDGHP\ LQ +H JUDGXDWHG IURP 7KH 8QLYHUVLW\ RI $ODEDPD LQ ZLWK D %60( DQG LQ ZLWK D 060( :KLOH DW WKH 8QLYHUVLW\ KH ZDV D PHPEHU RI 7DX %HWD 3L 3L 7DX 6LJPD $60( DQG $6+5$( ,Q DGGLn WLRQ :DOODFH ZDV D FRRS VWXGHQW ZLWK $ODEDPD 3RZHU &RPSDQ\ DW *UHHQH &RXQW\ 6WHDP 3ODQW QHDU 'HPRSROLV $ODEDPD ,Q :DOODFH MRLQHG &KHYURQ 86$ DW WKH 3DVFDJRXOD 5HILQHU\ LQ 3DVFDJRXOD 0LVVLVVLSSL +H HQMR\HG VHYHUDO DVVLJQPHQWV ZLWK &KHYURQ EHIRUH KH UHWXUQHG LQ WR VFKRRO DW WKH 8QLYHUVLW\ RI )ORULGD WR SXUVXH D 3K' LQ HQJLQHHULQJ PHFKDQLFV :DOODFH LV PDUULHG WR 5RE\Q DQ HGXFDWRU DQG 3K' VWXGHQW LQ VSHFLDO HGXFDWLRQ DW WKH 8QLYHUVLW\ RI )ORULGD

PAGE 117

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ D A&8AnT F ( 7D\ORU &KDLUPDQ 3URIHVVRU RI (QJLQHHULQJ 6FLHQFHV FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ • &.a e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

PAGE 118

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ ( :DOVK 3URIHVVRU RI (QJLQHHULQJ 6FLHQFHV 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI (QJLQHHULQJ DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 0D\ 'HDQ *UDGXDWH 6FKRRO

PAGE 119

81,9(56,7< 2) )/25,'$