高分辨率SAR散射机理与应用关键科学问题研究

项目名称： 高分辨率SAR散射机理与应用关键科学问题研究

项目编号： No.41331176

项目类型： 重点项目

立项/批准年度： 2014

项目学科： 天文学、地球科学

项目作者： 王超

作者单位： 中国科学院遥感与数字地球研究所

项目金额： 308万元

中文摘要： 高分辨率是SAR系统发展的最重要的趋势之一，大量的高分辨率SAR影像不仅为新型应用研究与发展提供了坚实的数据支持，也将推动和触发新的应用领域。. 如何充分利用高分辨率SAR能够反映地物细节结构的优势，提升SAR在传统应用上的精度，并引领发展高分辨率SAR的新型应用，从而促进我国高分辨率SAR应用的发展是本项目研究的推动力。因此，本项目将研究高分辨率SAR在极化、干涉的散射机理，发现高分辨率针对地物结构细节在散射强度、极化响应、相位相关性上的稳定性或变化规律；并在此基础上开展高分辨率SAR信息处理关键技术研究，重点突破高分辨率SAR针对典型地物的多特征参数分析与提取及时间序列分析。具体实现典型地物在两维空间上的精细分割、分类与提取、在三维空间上实现高可靠性、高精度的高度反演；在时间序列分析上实现对典型地物两维空间上的局部细节变化监测方法、以及三维空间上的微量沉降监测方法。

中文关键词： 雷达遥感;散射机理;多尺度分割与分类;三维信息反演;变化检测

英文摘要： High resolution is one of the most important trends for the development of synthetic aperture radar (SAR) systems. Numerous high resolution SAR images provide a firm data foundation for new researches and applications, and trigger studies in new areas..The aim of this proposal is to utilize the advantage of high resolution SAR images, which have the ability to reflect structure details of the ground objects, to improve the accuracy of traditional applications and to lead the revolution in innovative applications. Thus, this proposal will focus on the analysis of objects’ polarimetric and interferometric scattering properties in high resolution SAR images, specifically on the stability and variation of objects’ fine structures in scattering intensity, polarimetric response and phase interferometry. On the basis of the analysis, critical information processing techniques on high resolution SAR images will be focused, and research on multiple characterization extraction and time series analysis of typical ground objects will also be carried out. As a result, outputs of the proposal will be as follows: the new effective segmentation and classification methods for the high resolution SAR images, reliable and accurate height estimation algorithms for ground objects, and techniques deriving local micro variation and subsidence based on multi-temporary analysis of typical objects.

英文关键词： SAR remote sensing;Scattering mechanism;Multiscale segmentation;3D information inversion;Change detection