应用于图像分割的改进贪婪蛇算法  

Improved Greedy Snake Algorithm for Image Segmentation

作　　者：程乐[1,2] 周抒 宋艳红 杨晔[1] 卞曰瑭[4] CHENG Le;ZHOU Shu;SONG Yanhong;YANG Ye;BIAN Yuetang(Department of Computer Science and Communication Engineering,Huaian Vocational College of Information Technology, Huaian 223003,China;College of Computer and Information,Hohai University,Nanjing 210098,China;School of Computer and Software,Nanjing University of Information Science&Technology,Nanjing 210044,China;School of Business,Nanjing Normal University,Nanjing 210023,China)

机构地区：[1]淮安信息职业技术学院计算机与通信工程学院,江苏淮安223003 [2]河海大学计算机与信息学院,江苏南京210098 [3]南京信息工程大学计算机与软件学院,江苏南京210044 [4]南京师范大学商学院,江苏南京210023

出　　处：《济南大学学报（自然科学版）》2018年第3期212-217,共6页Journal of University of Jinan(Science and Technology)

基　　金：国家自然科学基金项目(71301078);江苏省高校自然科学基金项目(16KJB520049);淮安市自然科学研究计划项目(HAB201709);淮安信息职业技术学院创新基金项目(hxyc2015001)

摘　　要：针对现有的贪婪蛇算法存在的计算量大且不能很好地处理凹形图像的问题,提出一种改进贪婪蛇算法;该算法对原始图像进行灰度预处理,以提高原始图像锐化程度;通过重新设计能量函数中的图像力的计算方法,得到一种新的贪婪蛇算法能量函数,用于完成蛇素的初始收敛;使用一种贪婪收敛策略实现蛇素的最终收敛。图像分割实验验证了改进贪婪蛇算法的有效性,特别是在分割复杂凹形图像时效果较好。The existing greedy snake algorithm(GSA)had the problems of generous calculation and failing to treat the image segmentation of concave region.To solve these problems,an improved greedy snake algorithm(IGSA)was present-ed.The origin image was executed a pretreatment of the gray,which effectively enhanced the edged.By improving the image force of energy functional,a new energy functional of GSA was proposed.The snaxels completed the preliminarily convergence by the improved energy functional.A greedy approach was used for the ultimate convergence of snaxels.The results of experiment show that IGSA is efficient and good at processing the image segmentation of concave region.

关 键 词：贪婪蛇算法 蛇素 贪婪方法 凹形图 图像分割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]