2D cross entropy method for image segmentation based on artificial bee colony optimization

摘要

A kind of image segmentation method with two-dimensional cross entropy was proposed based on the artificial bee colony algorithm to overcome the large amount of calculation and long computing time. Firstly, the principle of two-dimensional cross entropy threshold segmentation was analyzed. Then, the bionic mechanism and searching optimization process of the artificial bee colony algorithm were analyzed, and the threshold segmentation method of two-dimensional cross entropy combined with artificial bee colony algorithm was proposed. Finally, typical image segmentation experiments by using the proposed method were performed and the results were compared with two-dimensional cross entropy exhaustive segmentation method and two-dimensional entropy segmentation method based on Particle Swarm Optimization (PSO). Experimental results show that the speed of the proposed method is ten times faster than the two-dimensional entropy exhaustive segmentation method respectively. Moreover, the threshold selection accuracy and running speed of the proposed method are both better than the threshold segmentation method of two-dimensional cross entropy based on PSO. Therefore, the image segmentation method of two-dimensional cross entropy based on artificial bee colony algorithm can quickly and efficiently resolve image segmentation problems.