Title :
Edge length ratios: an affine invariant shape representation for recognition with occlusions
Author :
Glauser, T. ; Bunke, H.
Author_Institution :
Bern Univ., Switzerland
fDate :
30 Aug-3 Sep 1992
Abstract :
The authors propose a set of local shape descriptors for polygons based on edge length ratios. These descriptors are invariant under any affine transformation, i.e. translation, scaling, rotation and projection from 3D to any 2D plane. Furthermore, they are complete in the sense that they allow the reconstruction of any polygon in 3D space given at least three consecutive vertices. The authors then consider the recognition of 2D polygonal objects under any affine transformation and in the presence of occlusion. they introduce a recognition procedure that is based on the matching of edge length ratios, using a simplified version of the standard dynamic programming procedure commonly employed for string matching
Keywords :
computerised pattern recognition; computerised picture processing; dynamic programming; affine invariant shape representation; affine transformation invariance; dynamic programming; edge length ratio matching; edge length ratios; occlusions; polygons; shape recognition; Constraint optimization; Convergence; Dynamic programming; Iterative algorithms; Iterative methods; Noise robustness; Noise shaping; Shape; Solid modeling;
Conference_Titel :
Pattern Recognition, 1992. Vol.I. Conference A: Computer Vision and Applications, Proceedings., 11th IAPR International Conference on
Conference_Location :
The Hague
Print_ISBN :
0-8186-2910-X
DOI :
10.1109/ICPR.1992.201594