An evolutionary image matching approach

Bee colony optimization (BCO) is a meta-heuristic technique inspired by natural behavior of the bee colony. In this paper, the BCO technique is exploited to tackle the shape matching problem with the aim to find the matching between two shapes represented via sets of contour points. A number of bees are used to collaboratively search the optimal matching using a proposed proximity-regularized cost function. Furthermore, the proposed cost function considers the proximity information of the matched contour points; this is in the contrast to that these contour points are treated independently in the conventional approaches. Experimental results are presented to demonstrate that the proposed approach is able to provide more accurate shape matching than the conventional approaches.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► We use bee colony optimization to perform image matching. ► We propose a proximity-regularized cost function to perform image matching. ► The proposed approach can provide more accurate image matching than conventional approaches.