结合蚁群算法的Snake模型的医学图像分割方法  

A novel image segmentation algorithm based on snake and artificial ant colony

作　　者：曹会志[1] 王晨[1] 罗述谦[1] 

机构地区：[1]首都医科大学生物医学工程学院,北京100069

出　　处：《北京生物医学工程》2007年第3期245-248,共4页Beijing Biomedical Engineering

基　　金：国家自然科学基金(60472020)资助

摘　　要：图像分割是医学图像处理中一个困难而又极其重要的课题。本文提出一种新的结合Snake模型和蚁群算法的图像分割算法。Snake模型是一种将目标的轮廓模型与图像特征相匹配的分割方法,而蚁群算法可以帮助人们理解生物系统的原理以及在机器人技术、计算机图形学等领域已经得到广泛应用。本文在结合已有Snake模型和蚁群算法优点的基础上,提出了新的图像蚁群分割模型。实验结果表明,本文提出的分割方法能够比较好的保留图像的细节信息,并具有一定的抗噪声能力。Segmentation is one of the most difficult tasks in digital image processing. This paper presents a novel segmentation algorithm, which based on snake model and artificial ant colonies Snakes are well known method for matching an object＇ s contour model to features in an image. As pointed out by some recent literatures, the self-organizations of ants were successfully used for understanding biological systems and in many applications in robotics, computer graphics, etc. Considering the features of Snakes and artificial ant colonies, we present an extended model applied in image segmentation. The simulated results show the efficiency of the new algorithm, which is able to preserve the detail of the object and is insensitive to noise.

关 键 词：活动轮廓模型 蚁群算法 医学图像 图像分割 

分 类 号：R318.04[医药卫生—生物医学工程]