参照失效物理分析的永久散射体提取过程

A New Method of Permanent Scatterers Detection Based on Analyzing Physics of Failure

在雷达干涉测量中,永久散射体(PS)是一类不受时间、空间基线和大气延迟影响的目标,如何从SAR图像中自动识别出有效的永久散射体是PS干涉系统中关键的环节之一。文章从不确定性量化理论的角度考虑失效物理分析问题,利用失效物理分析方法研究PS影像特征与环境的关系,对多幅SAR图像进行独立分量分析(ICA)和小波包分解,然后考虑永久散射体的高信噪比特性,依据图像能量最大的原则获取分离后的稳定成分,并进行分割处理得到PS点的分布图。实验表明,该方法能够有效识别出可靠的PS点。

In synthetic aperture radar interferometry, permanent scatterers （PSs） can be regarded as a series of measurement points which are not or less influenced by temporal and geometrical decorrelation and atmospheric delay. How to effectively detect and identify PS from SAR images is one of the crucial procedures in PS interferometric system. In this paper, the unascertained quantification theory is applied to failure analysis. And after discussing the relationship between PS and environment by analyzing physics of failure, a new method of permanent scatterers detection is presented by using the independent component analysis （ICA） and wavelet packet decomposition arithmetic. Considering the high signal-to-clutter radio （SCR） of PS, the stable component is determined according to the principle of maximum energy and the distribution map of PS is produced by threshold segmentation. The experimental results show that the proposed method can identify PS from SAR images effectively and reliably.