基于蚁群算法的最佳熵图像分割阈值方法  

Optimal Entropy Thresholding Image Segmentation Based on Ant Colony Optimization algorithm

作　　者：叶志伟[1] 常胜[2] 高山[3] 

机构地区：[1]湖北工业大学计算机学院,湖北武汉430068 [2]湖北民族学院生物科学与技术学院,湖北恩施445000 [3]武汉市规划土地管理信息中心,湖北武汉430014

出　　处：《湖北民族学院学报（自然科学版）》2007年第3期304-307,共4页Journal of Hubei Minzu University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(40271094)

摘　　要：最佳熵阈值是最常用的图像分割算法之一,但是需要大量的运算时间,限制了其实际的应用范围.蚁群算法是一种新兴的仿生进化算法,已成功的应用于大批组合优化问题的处理.将最大熵算法视为组合优化问题并引用蚁群算法加以处理,实验结果表明蚁群算法不仅可以实现最优阈值的确定,而且可以提高图像分割效率.The optimal entropy thresholding is one of the most popular algorithms in use of image segmentation, however,it needs a lot of computation time which limits its application. Ant colony optimization algorithm （ACO） was recently proposed algorithm, which has been successfully applied to solve many combinatorial optimization problems. On the analysis of optimal entropy, we are aware that threshold selection can be viewed as a combinatorial optimization problem. Thus, we introduce a new method to select image threshold automatically based on ACO algorithm. The performance of this algorithm is compared with optimal entropy, and experimental results show that ACO algorithm can not only determine the optimal threshold,but aoso improve the efficiency of image segmentation.

关 键 词：图像分割 阈值 最大熵 蚁群算法 

分 类 号：TN911.73[电子电信—通信与信息系统]