Evaluation of a Feature Extraction Method for Use in River Planform Analysis

MISSING IMAGE

Material Information

Title:
Evaluation of a Feature Extraction Method for Use in River Planform Analysis
Physical Description:
1 online resource (78 p.)
Language:
english
Creator:
Jones, Michal L
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Geography
Committee Chair:
Mossa, Joann
Committee Members:
Binford, Michael William
Waylen, Peter Robert

Subjects

Subjects / Keywords:
fluvial -- geomorphology -- gis -- planform -- river
Geography -- Dissertations, Academic -- UF
Genre:
Geography thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Image segmentation was tested as a replacement method for manual digitization of river channels from aerial imagery.  A simple t-test statistic was used to test the hypothesis that there would be no significant difference between the manual digitizing and image segmentation methods.  High spatial resolution orthorectified2008 imagery from the Florida Department of Transportation (FDOT) was used for comparison of the feature extraction methods. After several test iterations were run to determine the optimal segmentation parameters,each image was processed and converted to a shapefile for import into ArcMap10.  Once imported into ArcMap, the main river channel extracted from the image segmentation was overlaid with the channel extracted via manual digitization to visually compare the differences between the two, and clipped into 44 different statistical units.  The area of each channel was calculated within each statistical unit, and converted to a ratio.  The manual digitization method was used as the control method, assumed to have zero variability, and all 44 units were assigned a representative ratio of 1. The image segmentation units were each converted into a ratio by dividing the area of the image segmentation by the area of the manually digitized channel. The t-test statistical results of the first manual digitizing versus image segmentation comparison showed significant differences between the two techniques.  The image segmentation results with values farthest from 1 were quickly manually edited to be a better visual match to the river channel as seen in the imagery. New ratios were calculated and t-test statistic results of these ratios showed no significant difference between the two methods.  This demonstrated that the image segmentation method is an appropriate substitute for the manual digitizing method, producing repeatable results quickly and with little manual intervention.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Michal L Jones.
Thesis:
Thesis (M.S.)--University of Florida, 2013.
Local:
Adviser: Mossa, Joann.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-02-28

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0046029:00001


This item is only available as the following downloads:


Full Text

PAGE 12

Challenges with Mapping Rivers Inherent Versus Operational Errors et al

PAGE 13

Conversion Errors Registration Errors. Digitizing Errors. et al

PAGE 14

Spatial and Temporal Issues Spatial

PAGE 15

Temporal et al. et al

PAGE 16

A Matter of Scale catchment area river corridor

PAGE 17

local Other Sources of Error and Issues to Consider et al

PAGE 18

Image Segmentation et al

PAGE 19

et al

PAGE 21

Introduction

PAGE 22

Literature Review The Growing Need for GIS-Based Fluvial Studies

PAGE 23

Advantages of GIS-Based Studies et al et al et al et al

PAGE 25

et al Previous GIS-Based Studies on River Planform

PAGE 27

et al et al

PAGE 28

et al et al

PAGE 29

et al

PAGE 31

et al et al

PAGE 32

Study Area et al

PAGE 33

et al et al

PAGE 34

Methods Data Acquisition Geospatial Operations

PAGE 37

et al.

PAGE 38

Analysis and Results

PAGE 39

Discussion et al.

PAGE 40

Underestimation and Overestimation

PAGE 43

Manual Digitizing Method

PAGE 44

Time Advantages Conclusions

PAGE 45

et al

PAGE 70

70

PAGE 71

71

PAGE 72

72

PAGE 73

73 Discoveries Future Research

PAGE 74

74 Final Conclusions

PAGE 75

75 Geomorphology21 Professional Geographer 47 International Journal of Geographical Information Science 13 Variability in Stream Erosion and Sediment Transport Earth Surface Processes and Landforms 25 Water Resources Research 20 Environmental Management 25 Geomorphology 79 Regulated Rivers: Research & Management 13

PAGE 76

76 Hydrological Processes 12 Earth Surface Processes and Landforms 25 Geology28 Journal of the American Water Resources Association 37 Earth Surface Processes and Landforms 6 Restoration Ecology 3 Photogrammetric Record 16 Journal of Hydrology25 29 International Journal of Remote Sensing19 Geomorphology13 Applied Geography 17

PAGE 77

77 Applications of Geographic Information Systems in Hydrology and Water 1996 2009 IEEE International Conference on Image Processing, ICIP 2009 Proceedings Environmental Geology 41 American Journal of Science258 Water Resources Research 34 Geomorphology 21 Unpublished PhD Thesis. University of Stirling. Geomorphology34 Regulated Rivers: Research & Management 13 Regulated Rivers: Research & Manageme 16 International Journal of Applied Earth Observation and Geoinformation 1

PAGE 78

78