基于并行粒子群算法的Otsu双阈值医学图像分割  

The Otsu dual-threshold value method based on parallel particle swarm algorithm for medical image segmentation

作　　者：邬锡琴[1] 许良凤[1] 胡敏[1] 

机构地区：[1]合肥工业大学计算机与信息学院,安徽合肥230009

出　　处：《实验技术与管理》2011年第2期42-44,65,共4页Experimental Technology and Management

基　　金：肥工业大学博士基金资助项目(035022)

摘　　要：医学图像分割一直是医学影像分析领域的研究热点。由于粒子群优化(PSO)容易陷入局部极小,因此该算法用于搜索某些函数极值时精确度较低且稳定性较差。针对该问题,结合Otsu分割技术,提出了一种基于并行粒子群优化算法的Otsu双阈值医学图像分割算法。在该算法中,将粒子群体分成若干个子群体,进化在多个不同的子群中并行进行,避免单种群进化过程中出现的过早收敛现象,提高整个算法的收敛速度。实验结果表明,提出的分割算法与传统粒子群算法相比,不仅能够对图像进行准确的分割,而且具有更强的精确性和稳定性,其收敛速度明显优于基于单种群的粒子群算法的Otsu双阈值医学图像分割。The medical image segmentation is a hot topic in the field of medical images analysis.Because the particle swarm optimization（PSO） algorithm is easy to trap to local minima,the algorithm is sometimes inaccurate and instable when it is used in searching the best solutions of a function.To solve the problem,based on the Otsu segmentation method,the Otsu dual-threshold value method based on the parallel particle swarm algorithm for medical image segmentation is proposed.In the algorithm,the current particles are first divided into multi sub-population,the evolution is performed among different subgroups in parallel,and so this algorithm can avoid premature convergence of single-species evolutionary process,and improve the convergence efficiency of the algorithm.The results show that the presented algorithm can not only find better solutions,but also be more stable and accurate than that of the traditional particle swarm algorithm.The converge is improved more quickly than that of single-species particle swarm algorithm.

关 键 词：医学图像 Otsu双阈值法 粒子群算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]