遥感图象自动识别城市震害房屋——以2001年印度库奇地震和1976年唐山地震为例  

AUTOMATIC RECOGNITION OF DAMAGED TOWN BUILDINGS CAUSED BY EARTHQUAKE USING REMOTE SENSING INFORMATION:TAKING THE 2001 BHUJ, INDIA, EARTHQUAKE AND THE 1976 TANGSHAN, CHINA, EARTHQUAKE AS EXAMPLES

作　　者：柳稼航[1] 单新建[1] 尹京苑[2] 

机构地区：[1]中国地震局地质研究所地震动力学国家重点实验室 [2]上海市地震局,中国上海200062

出　　处：《地震学报》2004年第6期623-633,共11页Acta Seismologica Sinica

基　　金：上海市科学技术委员会 2 0 0 2年专项 (0 2 2 5 12 14 3);国家"十五"科技攻关项目 (2 0 0 1BA6 0 1B0 4 0 10 5 )共同资助

摘　　要：在高分辨率遥感影像上 ,完好房屋的表面呈现出均一的图象纹理特点 ,而对于倒塌房屋或半倒塌房屋 ,由于破坏截面比较粗糙 ,在图象上表现出斑块状的低灰度值区域 .用适当的灰度门限值截取图象亮度值时 ,在这些低灰度值区域会出现一系列互不连通的“洞” .利用震后一景遥感影像上的各区域洞的个数、洞面积与区域面积之比等统计信息 ,可将完好无损的房屋与损坏的房屋区分开来 ,从而达到破坏房屋与完好房屋自动识别的目的 .基于这些特点 ,本文提出了一种利用区域结构与纹理统计特性相结合进行损坏房屋自动识别的方法 ,并以 2 0 0 1年印度库奇 (Bhuj)地震 1m分辨率的iKonos卫星融合影像和 1976年我国唐山地震的黑白航空影像为例 ,利用这些区域结构和纹理的统计特性 ,进行倒塌房屋的自动识别 ,得到了较为满意的结果 .In the high-resolution images, the undamaged buildings generally show a natural textural feature, while the damaged or semi-damaged buildings always exhibit some low-grayscale blocks because of their coarsely damaged sections. If we use a proper threshold to classify the grayscale of image, some independent holes will appear in the damaged regions. By using such statistical information as the number of holes in every region, or the ratio between the area of holes and that of the region, etc. , the damaged buildings can be separated from the undamaged, thus automatic detection of damaged buildings can be realized. Based on these characteristics, a new method to automatically detect the damage buildings by using regional structure and statistical information of texture is presented in the paper. In order to test its validity, 1-m-resolution iKonos merged image of the 2001 Bhuj earthquake and grayscale aerial photos of the 1976 Tangshan earthquake are selected as two examples to automatically detect the damaged buildings. Satisfied results are obtained.

关 键 词：区域分析 震害识别 图象理解 库奇地震 唐山地震 

分 类 号：P315.9[天文地球—地震学]