Man-made Ground Collapse Detection Using High Resolution Aerial Image and Object-Based Classification: Example of Pearl River Delta

摘要：

In this paper, object-based image analysis method was used to detect the ground collapse sites using remote sensing images. Firstly, multi-scale image segmentation was performed on the 0.3-meter aerial image of study area and over tens of spatial, spectral, shape and texture features were extracted based on the segmented image objects. Then 4 optimized features for ground collapse classification was selected using generic algorithm (GA), which get the best fitness value in ground collapse classification.. After that, some on the spot land collapse sites were selected as cases sites and cased-based-reasoning(CBR) classification was applied on all the segmented image objects, from large scale to small scale. In the end, classification accuracy was evaluated over the whole study area. The average error of object-based CBR classification of ground collapse area is about 20.7％.

关键词： Ground collapse urban geology high resolution image object-based classification CBR(Case-Based Reasoning).

作者: Junping Qian Xiaozhan Zheng Shuisen Chen Ruihua Liu Jie Dou

作者单位: Guangzhou Institute of Geography, China Sun Yat-sen University, Guangzhou, China Guangzhou Institute of Geological Survey,China Guangzhou Institute of Geography, China

会议类型: 国际会议

会议名称: The Second International Workshop on Information Systems for Crisis Response and Management(ISCRAM CHINA 2007)(第二届国际危机响应与管理信息系统中国研讨会)

会议地点: 哈尔滨

会议语种:英文

页码: 110-114

在线出版日期: 2007-08-26（万方平台首次上网日期，不代表论文的发表时间）