Landslide susceptibility evaluation based on fuzzy support vector machine

The aim of this study was to apply and to verify the use of Fuzzy Support Vector Machine (FSVM) for Landslide susceptibility evaluation. The goal of the study includes four stages: preparation of a landslide inventory of the study area, extraction of influential factors of landslide, calculation of fuzzy membership values and evaluation of landslide susceptibility based on FSVM. The developed approach was tested in the Putian,Fujian,China. Geological data and satellite images were collected, processed, and constructed into a spatial database using GIS and image processing. The chosen factors that influence landslide occurrence were: slope angle, elevation, all from the topographic database; lithology taken from the geologic database; vegetation index value and wetness index value, from Landsat Thermatic Mapper (TM) satellite images. Each of the above-listed factors was classified and coded using a numerical scale corresponding to the physical conditions of the region. Using those factors, landslide susceptibility was evaluated by FSVM method, and the fuzzy membership values were calculated by combining frequency ratio and fuzzy operator. The results of the landslide susceptibility evaluation were verified using landslide location data. The validation results showed satisfactory agreement between the landslide susceptibility evaluation map and the existing data on landslide location.