基于Hlder指数和ICA的海域SAR图点目标检测(英文)  

Point Target Detection in Sea-Clutter SAR Image Based on Hlder Exponent and ICA

作　　者：李昱彤[1] 周越[1] 杨杰[1] 

机构地区：[1]上海交通大学图像处理与模式识别研究所,上海200240

出　　处：《光电工程》2008年第1期16-21,共6页Opto-Electronic Engineering

基　　金：航空科学基金(04F57004);国家自然科学基金资助项目(60675023)

摘　　要：本文将分形 Hlder 指数和信号分离相结合,利用独立成分分析技术(ICA, Independent Component Analysis),实现了海杂波 SAR 图的散斑抑制和点目标检测。首先,计算点态 Hlder 指数图,并提出二值模糊方法对其处理;接着使用 ICA 技术得到该图的基图像和独立成分;提出空间分离法,对独立成分进行分离,同时对基图进行对应分类,获得非噪声和噪声两个空间。最后在非噪声空间上重构图像。实验部分,将该算法与传统算法进行对比,证实了该算法的有效性和优越性。A novel approach is proposed for point target detection in sea clutter Synthetic Aperture Radar （SAR） image. Firstly, pointwise Hǒlder exponent of the original image is computed and proposed binary-fuzzy processing is applied on this HOlder image for target enhancement, and then, the processed image data is used as input data to Independent Component Analysis （ICA）. After convergence, based on the space separation theory we develop, original space of modified Hǒlder image is separated into two subspaces called clean-space and noise-space consisting of respective independent components and corresponding basis images. Finally, the detection result is obtained with construction on the ‘clean＇ subspace and clear extraction is obtained with simple morphological operation. As the results show, the proposed method is validated and point target is detected more efficiently compared with conventional methods.

关 键 词：点目标检测 合成孔径雷达图像 海杂波散斑抑制 Hǒlder指数 空间分离 

分 类 号：TN911.73[电子电信—通信与信息系统]