Affine invariant object shape matching using genetic algorithm with multi-parent orthogonal recombination and migrant principle

Genetic algorithm has served as a viable means for object shape matching. However, the success rate of correctly matching a pair of contours is generally low and inconsistent. In this paper a novel scheme based on the integration of two techniques: the migrant principle and multi-parent orthogonal recombination, are introduced to alleviate the problem. The proposed algorithm has been evaluated with object contours that are captured under good lighting condition, as well as images that have been contaminated with noise. In both cases experimental results revealed that the proposed method exhibited superior performance over its peers in terms of stability, speed, and success rate. © 2008 Elsevier B.V. All rights reserved.