模糊2-划分熵阈值法的快速迭代算法  

Fast iterated algorithm based on fuzzy 2-partition entropy for image thresholding

作　　者：曹利戈[1] 吴成茂[2] 

机构地区：[1]西安邮电学院后勤集团,陕西西安710121 [2]西安邮电学院电子与信息工程系,陕西西安710121

出　　处：《西安科技大学学报》2009年第6期769-774,共6页Journal of Xi’an University of Science and Technology

基　　金：国家自然科学基金项目(60572133);陕西省教育厅项目(06JK194)

摘　　要：基于模糊划分熵的图像阈值化分割方法因具有良好的分割性能已成为图像分割广泛应用的方法之一。但是,随其需要确定的隶属函数参数个数的增加,导致其搜索空间快速增大,不利用模糊划分熵阈值法在诸如机器视觉、目标检测等众多实时性要求较高场合的应用。提出了搜索模糊2-划分熵最佳分割阈值的快速迭代算法,使其计算复杂性从O(L3)降低为O(L).大量实验结果表明,提出的模糊划分熵阈值法所对应的快速迭代是可行的。hnage thresholding segmentation method based on fuzzy partition entropy has good segmentation performance and becomes a kind of extensive application method in image segmentation field. But the size of search space increases very rapidly when the number of parameters needed to be determined for the membership function increases. Traditional search method of selecting optimal segmentation thresholds based on fuzzy partition entropy is not suitable for the situation of high real - time such as machine vision, object detection etc. Fast iterative algorithm of searching optimal threshold based on fuzzy 2 -partition entropy is presented and its computation complexity reduces from O（ L^3 ） to O （L）. Many experimental results show that the fast iterative algorithm based on fuzzy 2 - partition entropy for 2 - level image thresholding is feasible.

关 键 词：图像分割 阈值法 模糊划分熵 隶属函数 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]