Ant colony matching: A curve evolution approach

In this paper we present a novel shape matching approach for 2D shapes. Finding correspondence for a huge number of points can be very complex. In order to resolve the complexity of matching between the huge number of points, we need to reduce the number of points incorporated in shape correspondence. Discrete Curve Evolution (DCE) proposed previously is selected to simplify the extracted contour. Then the correspondence between shapes is formulated as a Quadratic assignment problem (QAP) and resolved by using Ant Colony optimization (ACO). The experimental results demonstrate the effectiveness of our approach and give an intuitive shape similarity invariant to noise and distortions.