崔丽萍, 王晓青, 窦爱霞, 金鼎坚. 2016: 基于高分辨率合成孔径雷达影像建筑物成像几何结构的震害特征分析. 地震学报, 38(2): 272-282. DOI: 10.11939/jass.2016.02.011
引用本文: 崔丽萍, 王晓青, 窦爱霞, 金鼎坚. 2016: 基于高分辨率合成孔径雷达影像建筑物成像几何结构的震害特征分析. 地震学报, 38(2): 272-282. DOI: 10.11939/jass.2016.02.011
Cui Liping, Wang Xiaoqing, Dou Aixia, Jin Dingjian. 2016: Building damage analysis based on high resolution synthetic aperture radar imaging geometry. Acta Seismologica Sinica, 38(2): 272-282. DOI: 10.11939/jass.2016.02.011
Citation: Cui Liping, Wang Xiaoqing, Dou Aixia, Jin Dingjian. 2016: Building damage analysis based on high resolution synthetic aperture radar imaging geometry. Acta Seismologica Sinica, 38(2): 272-282. DOI: 10.11939/jass.2016.02.011

基于高分辨率合成孔径雷达影像建筑物成像几何结构的震害特征分析

Building damage analysis based on high resolution synthetic aperture radar imaging geometry

  • 摘要: 传统的利用震后单幅合成孔径雷达(SAR)影像对建筑物的震害特征分析大多基于街区范围, 很少基于其成像几何结构. 本文基于高分辨率SAR影像上的建筑物成像几何结构, 分析了建筑物单体的震害特点, 建立了利用距离向线性灰度累加的方法提取规则未倒塌建筑物的叠掩区和阴影区及倒塌建筑物的倒塌区, 并在此基础上进行各几何特征区域的纹理特征, 如同质度、 不相似度和熵的计算及其组合特征分析, 由此建立了基于SAR影像建筑物成像几何结构的震害分析方法. 采用该方法对2010年玉树MS7.1地震震后玉树县城区的高分辨率SAR影像进行分析, 结果表明: 叠掩、 阴影和二次散射亮线是进行建筑物震害解译的有效几何结构特征, 其中叠掩区和阴影区的影像纹理特征具有较好的震害识别能力; 与传统的简单特征统计方法相比, 考虑建筑物SAR影像成像几何结构的特征统计法, 可以显著提高建筑物的震害识别能力.

     

    Abstract: Traditional building damage detection using post-event synthetic aperture radar (SAR) imagery is usually performed based on street block area, few on imaging geometry structure (IGS). In this paper, the characteristics of individual damaged or undamaged buildings are firstly analyzed based on SAR IGS. Then a building damage analysis method based on SAR IGS is proposed by accumulating the SAR intensity in range direction in order to extract the layover and shadowing areas of undamaged buildings as well as the damaged areas of collapsed buildings. Next, we compute the texture features such as homogeneity, dissimilarity and entropy in these areas, and compare the composited texture features with undamaged and collapsed buildings. Finally, the post-earthquake airborne high resolution SAR data of Yushu MS7.1 earthquake are collected and analyzed by using the above method. The results show that layover, shadow and double bounce are effective features of SAR IGS to identify whether a building is damaged or not, among which the texture features in layover and shadow-ing areas perform well. What is more, using SAR IGS can apparently improve the recognition ability of damage detection compared with traditional methods using statistical features. The SAR IGS is of great significance to study and application in detecting collapsed buildings from high resolution SAR imagery.

     

/

返回文章
返回