首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于改进蚁群算法和指数熵算子的三类阈值图像分割
引用本文:安海涛.基于改进蚁群算法和指数熵算子的三类阈值图像分割[J].内蒙古农业大学学报(自然科学版),2012,33(2):182-184.
作者姓名:安海涛
作者单位:内蒙古电子信息职业技术学院实习实训中心,呼和浩特,010070
摘    要:针对多目标图像分割问题,采用了1种基于二维灰度直方图的三类阈值分割方法,将图像划分为暗、灰和亮3种不同的区域,分别给出了其模糊隶属度函数,引入概率分析,定义了基于指数熵算子的最大模糊熵准则,通过改进的蚁群算法,搜索最优模糊参数组合,以此确定图像的分割阈值。实验结果表明,该算法能快速、有效的分割图像。

关 键 词:图像分割  最大模糊熵  指数熵  蚁群算法

A THREE - LEVAL THRESHOLDING METHOD FOR IMAGE SEGMENTATION BASED ON IMPROVED ANT COLONY ALGORITHMS AND EXPONENTIAL ENTROPY OPERATOR
AN Hai-tao.A THREE - LEVAL THRESHOLDING METHOD FOR IMAGE SEGMENTATION BASED ON IMPROVED ANT COLONY ALGORITHMS AND EXPONENTIAL ENTROPY OPERATOR[J].Journal of Inner Mongolia Agricultural University(Natural Science Edition),2012,33(2):182-184.
Authors:AN Hai-tao
Institution:AN Har-tao(Practice Training Center,Inner Mongolia Electronic Information Vocational Technical College,Huhhot 010070,China)
Abstract:For multi-target image segmentation,a three-level thresholding method is presented based on 2D gray histogram,the image is partion to three parts fuzzily through defining three fuzzy membership functions,including dark,gray and bright part,the maximum fuzzy entropy is defined based on exponential entropy operator and probability analysis.According to the improved ant colony algorithms,the optimal combination of the fuzzy parameters is searched,and the optimal thresholds is determined.The experimental results show that the proposed method can segment the image effectively and fast.
Keywords:Image segmentation  maximum fuzzy entropy  exponential entropy  Ant colony algorithms
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号