基于区域相似性的活动轮廓SAR图像分割  

Region-Similarity Based Active Contour Model for SAR Image Segmentation

作　　者：孔丁科[1] 汪国昭[1] 

机构地区：[1]浙江大学数学系计算机图象图形研究所,杭州310027

出　　处：《计算机辅助设计与图形学学报》2010年第9期1554-1560,共7页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(60773179;60933008;60970079)

摘　　要：为了克服传统活动轮廓SAR分割模型高度依赖统计分布假定的缺点,结合基于成对相似性的图划分方法和几何活动轮廓模型的优点,提出了基于区域相似性的活动轮廓SAR分割模型.首先将原始图像过分割成同质子区域集;然后结合强度和纹理信息真实度量子区域的成对相似性,并以此定义能量泛函;最后利用基于过分割的规则化和快速曲线演化实现SAR图像的有效分割.真实SAR图像的实验结果表明,该模型能快速、准确地得到SAR图像的分割结果.Due to the fact that classical active contour models for SAR image segmentation are highly dependent on statistical distributions,a novel active contour model based on pairwise region similarity is proposed and used for SAR images segmentation.First,the image is over segmented into small homogenous regions.Then,a region similarity measure based on intensity and texture is defined and employed to construct an energy functional.Finally,an efficient regularization and curve evolution method based on over segmentation is enforced to improve the numerical accuracy and evolution efficiency.Experiments on SAR images show that our proposed model can both efficiently and accurately segment SAR images.

关 键 词：合成孔径雷达 图像分割 活动轮廓模型 区域相似性 过分割 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]