粒子群优化在图像最小误差阈值化中的应用  

Particle swarm optimization for image minimum error thresholding

作　　者：刘俊[1] 徐远远[2] 张跃飞[3] 郭进[4] 

机构地区：[1]西南交通大学招生就业处,成都610031 [2]河南省大气探测技术保障中心,郑州450003 [3]电子科技大学电子工程学院,成都610054 [4]西南交通大学信息科学与技术学院,成都610031

出　　处：《计算机应用》2008年第9期2306-2308,2311,共4页journal of Computer Applications

摘　　要：提出了一种基于粒子群优化(PSO)的图像最小误差阈值化方法。将粒子群优化算法应用于图像最小误差阈值化中,克服了常规最小误差阈值化计算量大的缺点。实验证明该算法能有效降低常规图像最小误差阈值化的计算量,与遗传算法相比,该方法有更好的收敛性和稳定性。A scheme for image minimum error thresholding based on Particle Swarm Optimization （PSO） was proposed in this paper. Minimum error thresholding is a popular technique for image segmentation, and our scheme applies PSO algorithm to image minimum error thresholding. This scheme solves the problem of large calculation burden of the regular minimum error thresholding method. The results we obtained demonstrate that our scheme can efficiently reduce calculation burden of the regular method and achieve better performance on convergence and robustness, compared with Genetic Arithmetic （GA）.

关 键 词：图像分割 粒子群优化 最小误差阈值化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]