基于M-S模型的三种图像分割算法的比较  

Comparison of three image segmentation algorithms based on M-S model

作　　者：党文静 李德权[1] 韦慧[1] DANG Wen-jing;LI De-quan;WEI Hui(College of Science, Anhui University of Science and Technology, Huainan Anhui 232001, China)

机构地区：[1]安徽理工大学理学院,安徽淮南232001

出　　处：《阜阳师范学院学报（自然科学版）》2017年第1期80-84,共5页Journal of Fuyang Normal University(Natural Science)

基　　金：国家自然科学基金项目(61472003;11601007)资助

摘　　要：M-S模型的水平集图像分割方法依赖于图像同质区域的全局信息,因而分割过程时间效率较低。为了提高计算效率,该方法在图像处理领域得到很多改进。本文在简化的M-S模型即C-V模型的基础上,讨论了现有3种改进分割演化算法,即:去掉C-V模型中的正则项;用||??取代狄拉克函数,使得方法具有更好的全局优化性;加入梯度局部项,使之适合处理弱边缘和边缘断裂的图像。最后,通过3个实例进一步验证了各算法的优劣性以及适用性范围。The level set image segmentation method of M-S model depends on the global information of image homogeneous region,thus the time efficiency in segmentation process is low.To improve the computational efficiency,this method has been improved by many researchers in the field of image processing.In this paper,the advantages and disadvantages of three kinds of improved segmentation evolutionary algorithms based on the C-V model is discussed:the algorithm based on the C-V model without the regularization term;the algorithm of replacing the Dirac function by|??|for the purpose of better global optimization;the algorithm of adding the local gradient term suitable for dealing with image of weak edges and edges fracture.Finally,three examples is presented to further illustrate the efficiency and the range of applicability of the algorithms.

关 键 词：M-S模型 C-V模型 水平集方法 图像分割 梯度 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]