一种新的水平集进化模型  

Novel level set evolution model

作　　者：李宏友[1] 汪同庆[1] 叶俊勇[1] 刘青[1] 

机构地区：[1]重庆大学光电技术及系统教育部重点实验室,重庆400030

出　　处：《仪器仪表学报》2008年第7期1365-1369,共5页Chinese Journal of Scientific Instrument

基　　金：国家科技支撑计划课题(2007BAG06B06)资助项目

摘　　要：针对目前图像分割领域许多水平集进化模型需要不断重新初始化水平集函数,或需要图像的梯度信息来约束进化的问题,提出了一种带距离约束项的基于亮度信息的水平集进化模型IMDC(intensity-based model with distance constraint)。该模型引入一个距离约束项作为内部能量来保证水平集函数始终不偏离符号距离函数(SDF),避免了进化过程中对水平集函数的不断初始化。同时,借鉴C-V模型的基本思想,采用图像的亮度信息而非梯度来构造模型的外部能量项,确保了零水平集曲线稳定地收敛于期望的图像特征点(如目标轮廓点)。实验结果表明,本文提出的模型不仅有效地克服了传统模型需重新初始化或无法应对弱边缘特征这两大问题,而且具备全局最优分割的能力和较强的抗噪性能。Most of current level set evolution models for image segmentation have to re-initialize the level set function constantly or require the gradient flow information to stop the evolution of the curve. To solve the problems, we propose a novel level set evolution model called IMDC （Intensity-based Model with Distance Constraint）, which consists of an internal energy term and an external energy term. The internal energy term includes a distance constraint that insures the deviation of the level set function from a signed distance function （SDF）. And the external energy term adopts the intensity instead of the gradient of the image to drive the curve on zero level set toward the desired image features, such as the object boundaries. Experimental results show that the proposed model efficiently avoids the reinitialization and overcomes the problem that the traditional models can not work well with the images with low gradient. Moreover, our model is able to acquire the global optimization of the segmentation and has a good anti-noise performance.

关 键 词：图像分割 水平集 几何主动轮廓模型 重新初始化 全局最优 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]