基于人工蜂群优化的二维最大熵图像分割.pdfVIP

2012 年 5 月 计算机工程 第 38 卷第9 期 Vo l.38 No.9 Computer Engineering May 2012 文章编号: 100伽-3428(2012)0弘~223--03 文献标识码.A 中图分类号. T阴11.73 ·图形图像处理· 基于人工蜂群优化的二维最大精图像分割 阿里木·赛买提t ，社培军E-Z ，柳思聪E (1.中国矿业大学江苏省资源环境信息工程理点实验室，江苏徐州 221116; 2. 南京大学卫星测绘技术与应用国家测绘地理信息局重点实验室，南京 210093) 摘要z 针对二维最大精罔像分割方法计算最大的问题，提出基于人工蜂群优化的二维援大销罔像分割算法.利用人工蝉群优化算法收敛 侠、选兔局部晨优、控制参数少等优点，将二维层大烧法最佳二维阕值视为.Q伴，蜜源，实现基于人工蜂群优化的二维层大销罔像分割。实 验结果表明，该方法的收敛速度较快、抗噪性较强。 关键词:图像分割;二续最大销;人工蜂群;粒子群优化;遗传算法;人工鱼群;遗传模拟退火算法 Maximum 2D Entropy Image Segmentation Based on Artificial Bee Colony Optimization AlimSamat , DU pei-jun ,2, LlU Si-cong (1. Jiangsu Key Laboraωry ofResources and Environmentallnfonnation Engineering, China University ofMining and Technology, Xuzhou 221116, China; 2. Key Laboratory for Satellite S町veying Technologies and Applications of State Administration of Surveying, Mapping and Geoinfonnatio口， Nanjing University, Nanjing 210093 , China) (AbstractJ Aiming at the problem oflarge computing in maximum 20 entropy based image segmentation method，由is paper proposes a maximum 20 entropy image segmentation algorithm based on artificial bee colony optimization. Artificial bee colony algorithm has certain advantage in conve鸣:en臼 speed， prevents loca1 optimization, and has few control p盯ameters. Using these advantages, the best 20 threshold of maximum 20 entropy method is considered as nectar, and artificial bee colony optimized maximum 20 entropy method is used ωsegment images. Experimental