Image registration by using statistical information

MISSING IMAGE

Material Information

Title:
Image registration by using statistical information
Physical Description:
1 online resource (114 p.)
Language:
english
Creator:
Lee, Jin-Seop
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mathematics
Committee Chair:
Chen, Yunmei
Committee Members:
Zhang, Lei
Rao, Murali
Shen, Li C
Ge, Jian

Subjects

Subjects / Keywords:
aos -- image -- information -- measure
Mathematics -- Dissertations, Academic -- UF
Genre:
Mathematics thesis, Ph.D.
Electronic Thesis or Dissertation
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )

Notes

Abstract:
Image registration has been an important subject in many areas including remote sensing, computer vision, and medical imaging. It has been intensively studied and many registration techniques have been developed. This dissertation focuses on the development of inverse consistent non-rigid image registration which are used, especially, in medical image area. In this dissertation, we first propose a novel variational model for mono-modal non-rigid image registration. This model uses a similarity measure derived from the maximum likelihood estimation of the residual with different local variances. By automatically updated different local variances at each iteration, the choice of the weight parameter balancing the regularization term and the similarity measure term becomes more flexible, and the accuracy and robustness are also improved. Secondly, we propose a new variational model for multi-modal non-rigid image registration, where the similarity measure is derived from R´enyi’s statistical dependence measure. By using this measure, the difficulty of estimation of the joint pdf, which is required in the mutual information based registration, can be avoided. Furthermore, the computation is simplified by using the theory of reproducing kernel Hilbert space. Thirdly, we propose a new variational model for inverse consistent non-rigid image registration. This model computes the forward and backward transformations that deform the source image to match the target image and vice versa, by minimizing a energy function that consists of the regularization term of the deformation fields, the error term of the alignment, and the inverse inconsistency cost term. By applying the inverse consistent registration, measurements or segmentations on one image can be precisely transferred to the other, which is especially important for medical treatment such as image-guided radiation therapy and surgery. Lastly, we propose a new variational model for bidirectional inverse consistent multi-modal non-rigid image registration. That model deforms the source image and the target image simultaneously and minimizes the disparity between the deformed source image and the deformed target image. This method can handle relatively larger deformation and is good for parallel computing.
Statement of Responsibility:
by Jin-Seop Lee.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
General Note:
Description based on online resource; title from PDF title page.
General Note:
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.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
General Note:
Adviser: Chen, Yunmei.
General Note:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-05-31

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Jin-Seop Lee. 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.
Classification:
lcc - LD1780 2013
System ID:
UFE0045405:00001