Retinal Fundus Image Registration via Vascular Structure Graph Matching
Motivated by the observation that a retinal fundus image may contain some unique geometric structures within its vascular trees which can be utilized for feature matching, in this paper, we proposed a graph-based registration framework called GM-ICP to align pairwise retinal images. First, the retin...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2010-01-01
|
| Series: | International Journal of Biomedical Imaging |
| Online Access: | http://dx.doi.org/10.1155/2010/906067 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832554911015895040 |
|---|---|
| author | Kexin Deng Jie Tian Jian Zheng Xing Zhang Xiaoqian Dai Min Xu |
| author_facet | Kexin Deng Jie Tian Jian Zheng Xing Zhang Xiaoqian Dai Min Xu |
| author_sort | Kexin Deng |
| collection | DOAJ |
| description | Motivated by the observation that a retinal fundus image may contain some unique geometric structures within
its vascular trees which can be utilized for feature matching, in this paper, we proposed a graph-based registration
framework called GM-ICP to align pairwise retinal images. First, the retinal vessels are automatically detected and
represented as vascular structure graphs. A graph matching is then performed to find global correspondences between
vascular bifurcations. Finally, a revised ICP algorithm incorporating with quadratic transformation model is used at
fine level to register vessel shape models. In order to eliminate the incorrect matches from global correspondence
set obtained via graph matching, we proposed a structure-based sample consensus (STRUCT-SAC) algorithm. The
advantages of our approach are threefold: (1) global optimum solution can be achieved with graph matching; (2)
our method is invariant to linear geometric transformations; and (3) heavy local feature descriptors are not required.
The effectiveness of our method is demonstrated by the experiments with 48 pairs retinal images collected from
clinical patients. |
| format | Article |
| id | doaj-art-74ecb4efe44549c1b0a5afcf71c84c56 |
| institution | Kabale University |
| issn | 1687-4188 1687-4196 |
| language | English |
| publishDate | 2010-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | International Journal of Biomedical Imaging |
| spelling | doaj-art-74ecb4efe44549c1b0a5afcf71c84c562025-02-03T05:50:14ZengWileyInternational Journal of Biomedical Imaging1687-41881687-41962010-01-01201010.1155/2010/906067906067Retinal Fundus Image Registration via Vascular Structure Graph MatchingKexin Deng0Jie Tian1Jian Zheng2Xing Zhang3Xiaoqian Dai4Min Xu5School of Electronic Engineering, Xidian University, Xi'an, Shanxi 710071, ChinaSchool of Electronic Engineering, Xidian University, Xi'an, Shanxi 710071, ChinaInstitute of Automation, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Automation, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Automation, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Automation, Chinese Academy of Sciences, Beijing 100190, ChinaMotivated by the observation that a retinal fundus image may contain some unique geometric structures within its vascular trees which can be utilized for feature matching, in this paper, we proposed a graph-based registration framework called GM-ICP to align pairwise retinal images. First, the retinal vessels are automatically detected and represented as vascular structure graphs. A graph matching is then performed to find global correspondences between vascular bifurcations. Finally, a revised ICP algorithm incorporating with quadratic transformation model is used at fine level to register vessel shape models. In order to eliminate the incorrect matches from global correspondence set obtained via graph matching, we proposed a structure-based sample consensus (STRUCT-SAC) algorithm. The advantages of our approach are threefold: (1) global optimum solution can be achieved with graph matching; (2) our method is invariant to linear geometric transformations; and (3) heavy local feature descriptors are not required. The effectiveness of our method is demonstrated by the experiments with 48 pairs retinal images collected from clinical patients.http://dx.doi.org/10.1155/2010/906067 |
| spellingShingle | Kexin Deng Jie Tian Jian Zheng Xing Zhang Xiaoqian Dai Min Xu Retinal Fundus Image Registration via Vascular Structure Graph Matching International Journal of Biomedical Imaging |
| title | Retinal Fundus Image Registration via Vascular Structure Graph Matching |
| title_full | Retinal Fundus Image Registration via Vascular Structure Graph Matching |
| title_fullStr | Retinal Fundus Image Registration via Vascular Structure Graph Matching |
| title_full_unstemmed | Retinal Fundus Image Registration via Vascular Structure Graph Matching |
| title_short | Retinal Fundus Image Registration via Vascular Structure Graph Matching |
| title_sort | retinal fundus image registration via vascular structure graph matching |
| url | http://dx.doi.org/10.1155/2010/906067 |
| work_keys_str_mv | AT kexindeng retinalfundusimageregistrationviavascularstructuregraphmatching AT jietian retinalfundusimageregistrationviavascularstructuregraphmatching AT jianzheng retinalfundusimageregistrationviavascularstructuregraphmatching AT xingzhang retinalfundusimageregistrationviavascularstructuregraphmatching AT xiaoqiandai retinalfundusimageregistrationviavascularstructuregraphmatching AT minxu retinalfundusimageregistrationviavascularstructuregraphmatching |