Quantitative Analysis of Natural Calamities for Disaster Management through Machine Learning Techniques Using Satellite Data

Geographic information and Remote sensing systems have been used in recent years by disaster platforms for calamity warning and preparedness for earthquakes, landslides, and floods to aid in the assessment and management of catastrophic risk. In the realm of technology, it has also received a lot of attention. It would not be possible to use sensory data without the appropriate technology for organizing enormous amounts of data and obtaining information from numerous sources, such as maps or measurement channels. This study uses machine learning techniques to locate and evaluate locations damaged by landslides, earthquakes, floods, avalanches, and wildfires. After applying filters to improve the quality of the images, the images are classed using both supervised and unsupervised methods after being segmented using a thresholding technique. To analyze devastation, images from before and after disasters are collected from MRSAC Nagpur and processed with Python-based tools, ArcGIS, ERDAS, and QGIS.