Abstract A relaxation method based on patterns of local features is used to find matches between pairs of images or subimages that differ in position or orientation. A local operator is applied to the two images to detect two sets of “corners” C1 …, Cm and D1, …, Dn, each of them characterized by position, orientation, contrast and “sharpness” (of the angle). For each pair (Ci, Dj), a figure of merit is computed and a relaxation process is used to iteratively adjust these figures of merit, based on the merits of other pairs in approximately corresponding positions. After a few iterations of this process, “good” matches (pairs having much better merit than their next best choices) are clustered, yielding sets of transformation parameters (shift vectors or rotation angles) under which many corners correspond. This method has yielded good results for TV images of objects such as tools and industrial parts, as well as for aerial images of terrain.
[1]
Robert C. Bolles,et al.
Parametric Correspondence and Chamfer Matching: Two New Techniques for Image Matching
,
1977,
IJCAI.
[2]
Marc Berthod,et al.
Scene labeling: an optimization approach
,
1980,
Pattern Recognit..
[3]
Paul Beaudet,et al.
Rotationally invariant image operators
,
1978
.
[4]
Azriel Rosenfeld,et al.
Gray-level corner detection
,
1982,
Pattern Recognit. Lett..
[5]
Andrew K. C. Wong,et al.
Resolution-Dependent Information Measures for Image Analysis
,
1977,
IEEE Transactions on Systems, Man, and Cybernetics.
[6]
Hans P. Morevec.
Towards automatic visual obstacle avoidance
,
1977,
IJCAI 1977.
[7]
Azriel Rosenfeld,et al.
Point pattern matching by relaxation
,
1980,
Pattern Recognit..
[8]
Julius T. Tou,et al.
Pattern Recognition Principles
,
1974
.