Robust detection of skewed symmetries by combining local and semi-local affine invariants

Affine-invariant feature vector (Ip and Shen Image Vision Comput. 16 (2) (1 998) 135-146). that captures both local and semi-local geometric features a round each point of the object boundary is applied here for the detection o f skewed symmetries. Based on the affine-invariant shape representation, th e problem of detecting symmetry axes has been formulated as a problem of de tecting lines, with known orientations, in a local similarity matrix of an object. Since the feature vector extracts sufficient local and semi-local s hape information For every point along the object boundary, the process of checking symmetric point pairs is thus robust against both noises and defor mations. Moreover, our technique is able to detect all the local reflection al symmetries contained in the object. Various experimental results have sh own the robustness and effectiveness of our method in detecting skewed symm etries from both self-symmetric objects and generalized objects. (C) 2001 P attern Recognition Society, Published by Elsevier Science Ltd. All rights r eserved.