Image Segmentation Based on Improved Genetic Algorithm

摘要：

An evolutionary image segmentation method by genetic algorithm (GA) is presented.GA has global search abilities but it may suffer from premature convergence on multi-modal problems.This is due to the decrease of populations diversity in search space that leads to fitness stagnation.An improved GA (IGA) is proposed and the mutation rate in IGA is controlled by the populations diversity which can balance the populations diversity and have more chances located in the region of global solutions.The effectiveness of IGA is shown by the experiment of image segmentation,the results show that IGA has great advantage of convergence speed and success ratio over GA.

关键词： Genetic Algorithm global optimization image segmentation

作者: Xu Ling

作者单位: School of Information technology,Jiangnan University,Wuxi,Jiangsu,214122 Lake Tai College,Jiangnan University,Wuxi,Jiangsu,214064

会议类型: 国际会议

会议名称: 2008年国际电子商务、工程及科学领域的分布式计算和应用学术研讨会(2008 International Symposium on Distributed Computing and Applications for Business Engineering and Science)

会议地点: 大连

会议语种:英文

页码: 269-273

在线出版日期: 2008-07-27（万方平台首次上网日期，不代表论文的发表时间）