A fusion method of geometric and topological features for boundary-based shape matching and retrieval

In this paper, an efficient method for boundary-based shapes matching and retrieval in the presence of occlusion is presented. The geometric and topological information of boundary curves are encoded in the form of longest common sub-curves (LCS) graphs and their similarity is estimated by graph matching. B-spline is used for approximating the original boundary, then inflection points are detected to split such a Bspline to convex/concave segments. The characteristic string is constructed based on these segments´ canonical frame. After LCS candidates are found, their graphs, which are constructed using their segments as vertices and the weighted walkthrough (WW) between two segments as edges, are compared to obtain the optimal match. The experimental results and comparisons demonstrate that the proposed method outperforms traditional boundary-based methods in shape matching and it enhances the quality of inexact shape retrieval