Disaster Impact Analysis Uses Land Cover Classification, Case study: Petobo Liquefaction

Abstract

Analysis of changes in the conditions of an area can be done through satellite image analysis. This study utilizes the classification of satellite imagery to determine the impact of disasters and liquefaction disaster recovery efforts in the Petobo region, Palu, Central Sulawesi. The deep learning approach, namely Convolutional Neural Network (CNN) and CNN combined with ResNet as the Transfer Learning model, were selected as classification methods that would be compared in determining the approach with the best performance. The classification of satellite imagery is mapped into two main classes, namely natural land cover and artificial land cover. This research subsequently succeeded in mapping land cover changes that occurred as a result of liquefaction disasters and recovery efforts that have been carried out with promising performance