一种综合多特征的全极化SAR图像分割方法  

An Integrated Multi-feature Segmentation Method of Polarimetric SAR Images

作　　者：陈启浩[1] 刘修国[1] 陈奇[1] 

机构地区：[1]中国地质大学(武汉)信息工程学院,湖北武汉430074

出　　处：《武汉大学学报（信息科学版）》2014年第12期1419-1424,共6页Geomatics and Information Science of Wuhan University

基　　金：国家自然科学基金资助项目(41301477);中国博士后科学基金资助项目(2012M521497);武汉市学科带头人计划资助项目(201271130443);中国地质大学(武汉)中央高校基本科研业务费专项资金资助项目~~

摘　　要：提出了一种综合利用极化特征、统计特征和几何形状特征的全极化SAR图像分割方法。该方法采用分形网络演化算法思想,基于相干矩阵Pauli分解构建对象间的极化特征相似性准则,根据相干矩阵的Wishart分布假设构建对象间的统计特征相似性准则;制定对象合并过程中多特征的综合策略,通过极化特征增强及权重调整统一各类特征异质度的水平,最终建立全极化SAR图像多特征综合分割流程。实验结果表明,该方法能有效抑制斑点噪声,地物边界分割准确,特别是对具有均质纹理的农田、水体等分割效果较好。In this paper,we proposed a new segmentation method which integrates polarimetric,statistical distribution and geometric shape features of polarimetric SAR image based on the Fractal Network Evolution Algorithm（FNEA）.Firstly,the similarity criterion of polarimetric features between adjacent objects was acquired based on the Pauli decomposition,while the similarity criterion of statistical feature was constructed via the coherency matrix Wishart distribution hypothesis.Secondly,a multi-feature integration strategy in objects merging was established,and polarimetric feature values were stretched in advance to make the heterogeneities of polarimetric,statistic distribution and shape features between adjacent objects to be at the similar level.Then,an integrated multi-feature segmentation flow was built according to the above processes.Lastly,this method was verified with RADARSAT-2image of Altona and L band ESAR image of Oberpfaffenhofen,suggesting that it can effectively reduce the speckle effects and obtain accurate segmentation results,especially in the homogeneous texture areas like farmland and lakes.

关 键 词：全极化SAR 分割 多特征综合 分形网络演化算法 

分 类 号：P237[天文地球—摄影测量与遥感] TN957.52[天文地球—测绘科学与技术]