引力搜索算法在多阈值图像分割中的应用研究

万仁远1; 侯爱莲2

引力搜索算法在多阈值图像分割中的应用研究

引用本文：万仁远1，侯爱莲2.引力搜索算法在多阈值图像分割中的应用研究[J].计算技术与自动化,2018,(2):105-109

摘要点击次数: 824

全文下载次数: 0

作者	单位
万仁远1，侯爱莲2	（1.陕西铁路工程职业技术学院电气与信息工程系，陕西渭南 710042；2.中国人民银行长沙中心支行，湖南长沙 410005）

中文摘要:针对求解高维阈值图像分割计算复杂的问题，提出了一种基于引力搜索算法的多阈值图像分割方法，该方法以大津法（Otsu）设计为适应度函数，利用引力搜索算法快速搜索得到待分割图像的最优阈值，然后根据最优阈值进行图像分割。结合人眼视觉可知，引力搜索算法能够结合应用于图像分割，且能取得较好的效果。实验测试结果表明，该方法与布谷鸟算法、人工蜂群算法比较，引力搜索算法的收敛速度更快，寻优的阈值质量较高。

中文关键词:引力搜索算法图像分割多阈值 Otsu法

Research on Application of Gravitational Search Algorithm in Multi Threshold Image Segmentation

Abstract:In order to solve the complex problem of image segmentation of high dimensional threshold，a multi-threshold image segmentation method based on gravity search algorithm is proposed，the method designs the fitness by Otsu， and uses the favorable performance of gravitational search algorithm to fast and accurately find the optimal threshold of the image to be segmented.Then the image segmentation is carried out according to the optimal threshold.Combining with human vision，the gravitational search algorithm can be used in image segmentation and can achieve better results.Experiment results show that compared with cuckoo search algorithm and artificial bee colony algorithm， the performance of gravitational search algorithm is optimal in terms of its convergence speed and better quality threshold.

keywords:gravitational search algorithm image segmentation multi-threshold Otsu method

查看全文 查看/发表评论 下载pdf阅读器