一种基于图像边缘特征的SAR斑点滤波算法  

A new SAR image filter method based on edge detection

作　　者：曹云刚[1] 严丽娟[2] 郑泽忠[1] 

机构地区：[1]西南交通大学土木工程学院测量工程系,成都610031 [2]成都理工大学地质灾害防治与地质环境保护国家重点实验室,成都610059

出　　处：《测绘科学》2010年第3期165-166,120,共3页Science of Surveying and Mapping

基　　金：成都理工大学地质灾害防治与地质环境保护国家重点实验室开放基金项目(GZ2006-09)

摘　　要：合成孔径雷达图像上的斑点噪声会阻碍图像的校准、检验、解译和应用,因此斑点噪声的去除是雷达图像处理的一个重要环节。通常的去噪方法在抑制噪声的同时也使得图像的边缘模糊,细节特征损失。理论和实践表明去除图像的斑点噪声和保持图像的边缘信息是无法同步实现的。为了在去除噪声和保持边缘特征之间进行折衷,本文基于MROA边缘检测算法以及均值滤波算法,提出了一种基于图像边缘特征的斑点噪声滤波算法。实验结果表明该算法能够在平滑图像的同时有效保持边缘特征信息。Speckles in Synthetic Aperture Radar （SAR） images disturbs SAR images calibration, inspection, interpretation and application, thus speckle reducing is an important step in SAR images processing. Ordinary methods destroy edges information while reducing speckles. In fact, the experiments indicate that there is always a kind of tradeoff between smoothing out speckle and preserving the useful information. In this context, a SAR filter method based on edge detection result was proposed in this paper. With this method, modified ratio of average （MROA） algorithm for edge detection and mean filter algorithm were used for smoothing out speckle in non-edge area. Experiment result indicated that this method reduces speckles effectively while preserving image edges.

关 键 词：合成孔径雷达 斑点噪声 边缘检测 滤波 

分 类 号：TP722[自动化与计算机技术—检测技术与自动化装置] TP751[自动化与计算机技术—控制科学与工程]