文章基本信息

标题：Image Segmentation Algorithm Based on Improved Ant Colony Algorithm
本地全文：下载
作者：Xumin Liu ; Xiaojun Wang ; Na Shi 等
期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2014
卷号：7
期号：3
页码：433-442
DOI：10.14257/ijsip.2014.7.3.35
出版社：SERSC
摘要：In the process of image segmentation, the basic ant colony algorithm has some disadvantages, such as long searching time, large amounts of calculation, and rough image segmentation results. This paper proposes an improved ant colony algorithm. Applying different transfer rules and pheromone update strategies to different regions of an image, including background, target, edge and noise, we develop a highly adaptive image segmentation method with high edge detection accuracy and high algorithm implementation efficiency. In the initial stage of image segmentation, we apply the idea of fuzzy clustering, which enables ants to gather quickly to the edge in the background and the target area of the image. In the later stage of image segmentation, we introduce an edge search strategy in the edge area. A following experiment shows that this developed image segmentation method can split the target more quickly and accurately.
关键词：Ant Colony Algorithm; Image segmentation; Fuzzy Clustering; Edge search ; strategy