Landslide risk analysis using artificial neural network model focussing on different training sites

摘要: This paper presents landslide hazard and risk analysis using remote sensing data, GIS tools artificial neural network model. Landslide locations were identified in the study area from interpretation of aerial photographs field surveys. Topographical geological data satellite images collected, processed, constructed into a spatial database image processing. These factors used with to analyze hazard. Each factor’s weight was determined by back-propagation training method. Then indices calculated trained weights, map created tools. verify results maps compare them. The verified location compared model all cases. accuracy observed 83, 72, 82, 79 81% for sites 1, 2, 3, 4 5 respectively. efficiently large volume proved be an effective tool analysis. Further, has been performed DEM, distance zone, land cover damageable objects at risk. DEM delineate catchments served as mask extract highest zones area. Risk produced overlying techniques GIS. information could estimate population, property existing infrastructure like transportation network.