基于蚁群算法和模糊最大熵的测量靶面图像的分割方法  

A Segmentation Method of Target Image Based on Ant Colony Algorithm and Fuzzy Maximum Entro

作　　者：王孝贤 郭健 苟旭辉 关丽敏[2] 黄鹤[2] 王晓艳[2] Wang Xiaoxian;Guo Jian;Gou Xuhui;Guan Limin;Huang He;Wang Xiaoyan(Highway Project Construction Department of Weinan,Weinan,714000,China;School of Electronic&Control Engineering Chang’an University Xi’an 710064,China)

机构地区：[1]渭南市公路工程建设处,渭南陕西714000 [2]长安大学电子与控制工程学院,西安陕西710064

出　　处：《现代科学仪器》2018年第2期63-67,共5页Modern Scientific Instruments

基　　金：陕西省交通运输厅科技项目(17-16K;17-33T)。

摘　　要：针对特殊环境下获取的测量靶面图像的特征和系统的性能要求提出了一种适合测量靶面图像快速准确分割的模糊指数熵的多阈值图像分割方法。该方法首先将靶面图像分成若干个模糊集合,再运用最大模糊熵准则确定对应的隶属函数,并通过引入指数熵避免香农熵的缺陷,最后,结合蚁群优化算法的优点,借助蚁群优化算法快速准确的搜索图像分割的最优分割阈值,从而实现了对测量标尺靶靶面图案的快速准确分割,为后续标尺靶图像高精度测量提供了良好的保障。通过仿真和试验结果证明,该方法较传统的最大类间方差法分割速度提高了大约50%,而且分割效果有明显的改善,因此该方法能够快速的从特殊环境下获取的供测量的标尺靶面图像中分割出标尺图案,分割效果良好且速度快,满足工程要求。Considering the rule target image feature obtaining in special environment and the requirement of system,a novel high speed multi-threshold image segmentation algorithms that fitting the target segment have been designed.In the algorithms the rule-target image has been part into several fuzzy aggregates firstly.Then the fuzzy attached function can be established according as the maxim fuzzy entropy principle,at the same time the index entropy in order to avoid the shortcoming of Shanon entropy.Lastly the optimum threshold value be searched quickly and exactly help by the Ant Colony Optimization(ACO)algorithm.So the target image can be segment quickly and exactly which ensure the image measurement afterwards.Test and simulation shows these algorithms can segment the rule image of rule target that provide to measurement obtaining from special environment,the time consumed of the new algorithm is almost as 50%as the time consumed by the Maximum Between-Class Variance(OSTU),At the same time the segmenting result finely as well as saving time so it can meet the engineer request.

关 键 词：图像分割 模糊熵 蚁群优化 最大类间 最优组合 

分 类 号：TN911.7[电子电信—通信与信息系统]