利用多尺度小波变换改进PS点目标探测

如何从SAR图像中自动识别出有效的永久散射体（PS）是PS干涉系统中的关键环节之一。小波多尺度积能够反映目标信号与噪声在各频带之间的关联性,可以通过小波多尺度积来凸显目标。小波模极大值法通过搜索尺度空间中的模极大值,并利用信号和噪声不同的尺度特性来定位和识别目标。文章对这两种识别方法进行理论分析和实验对比,使用南京地区ENVI—SAT卫星的单视复图像数据进行了PS点的提取。实验结果表明：基于小波变换的多尺度PS点识别方法能够有效地降低雷达噪声的影像,可以在各种复杂的背景情况下有效地识别PS点。其中,小波模极大值法具有更为稳健的抗噪能力。

Multiscale Wavelet Transform for PS Point Target Detection

How to effectively detect and identify permanent scatterers(PS)from SAR images is one of the crucial procedures in PS interferometric system.In this paper,a new method of PS detection is presented by using the Wavelet Multi-scale Product and Wavelet Modulus Maxima according to the PS targets'features and the characteristics of speckle noise in SAR image.This paper analyzes the theory of wavelet multiscale product and wavelet modulus maxima and its suitability for identifying PS points respectively.Using 4 SAR images over Nanjing from satellites ENVISAT of European Space Agency,both of the methods have been validated for PS detection.The testing results show that both algorithms can identify PS from SAR images effectively and reliably,of which the method using wavelet modulus maxima has stronger antinoise ability.