The quantification of factors for landslide is very complex. Under different circumstances, the contribution of the same factor to landslide may be quite different or even opposite. Based on an analysis of the specific landslide data obtained from Yongjia County, the authors put forward a quantification method for building Neural Network, which is based on the distribution of landslide samples. The results from several estimation models were compared with each other. It is proved that the quantification method advanced by the authors and the Neural Network model based on Supported Vector Machine are the best means. The correctness is up to 84.2%, which is satisfying. According to the estimation result, the quantification solution and the estimation model based on it are very useful.
张重, 沈晓华, 邹乐君, 吴文渊, 苏楠, 孔凡立.
滑坡危险性评价模型中的量化方式研究——以永嘉县为例[J]. 国土资源遥感, 2010, 22(3): 16-20.
ZHANG Chong, CHEN Xiao-Hua, ZOU Le-Jun, WU Wen-Yuan, SU Nan, KONG Fan-Li. The Quantification Method in the Estimation Model for Landslide Danger: a Case Study of Yongjia County. REMOTE SENSING FOR LAND & RESOURCES, 2010, 22(3): 16-20.
[3]Chang K T,Liu J K. Landslide Features Interpreted by Neural Network Method Using a High-resolution Satellite Image and Digital Topographic Data [C]∥Proceedings of 20th ISPRS Congress. Istanbul:2004:574-579.
[4]Biswajeet Pradhan,Saro Lee. Utilization of Optical Remote Sensing Data and GIS Tools for Regional Landslide Hazard Analysis Using an Artificial Neural Network Model[J]. Earth Science Frontiers,2007, 14(6):143-152.