Image multi-level thresholds based on PSO with Von Neumann neighborhood
基于冯诺依曼邻居的粒子群多阈值分割算法*

:As for image multilevel thresholding, this paper proposed a novel algorithm. With modified Otsu method, the algorithm could get good thresholds by correctly finding the valley of image histogram. Based on particle swarm optimization with Von Neumann neighborhood, the algorithm improved the optimization performance. Also, described a way to build Von Neumann structure and compared its performance with global topology. Experimental results show that the proposed algorithm is promising and outperforms some existing techniques.