| Technology and Methodology |
|
|
|
|
|
A ROBUST MATCHING ALGORITHM BASED ON MATCHING SUPPORT |
| QIN Xu-wen1, GUAN Xiao-ping2, GUAN Ze-qun2 |
1. China Aero Geophysical Survey and Remote Sensing Center for Land and Resources, Beijing 100083, China;
2. School of Information Engineering in Remote Sensing, Wuhan University, Wuhan 430079, China |
|
|
|
|
Abstract This paper proposes a new mathematical model for defining the matching support, which can remarkably improve the stability of matching. First, the interesting points are extracted by Harris algorithm, and the candidate matches can be found through correlation. Then the matching support can be calculated according to length distortion, direction distortion, gray distortion, and double faces constraint. Finally, relaxation is used to eliminate ambiguity of the matches. The experiment proves that, with this algorithm, we can establish relatively robust results characterized by one-to-one correspondence.
|
|
Issue Date: 02 August 2011
|
|
|
| [1] Zhang Z, Derche R, Faugeras O,et al.A robust technique for matching two uncalibrated images through the recovery of the unknown epipolar geometry［J］. Artificial Intelligence Journal ,1995,78:87-119.[2] 吴晓良.影像匹配的松弛途径［D］.武汉:武汉测绘科技大学 ,1993.[3] 张祖勋,张剑清.数字摄影测量［M］.武汉: 武汉测绘科技大学出版社 ,1996.[4] Levy M.A new theoretical approach to relaxation, application to edge detection[A].Pattern Recognition,9th International Conference［C］.1988:208-212.[5] Gold S, Rangarajan A.A graduated assignment algorithm for graph matching［J］.IEEE Tans. Patt. Anal. Mach. Intel. 1996,18(4):377-388.[6] Haili Chui, Anand Rangarajan.A new Algorithm for Non-Rigid Point Matching[A].IEEE Conference on Computer Vision and Pattern Recognition［C］.2000. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
2003,
Vol. 15
