AI-Driven Approaches to Landslide Predictionwith Satellite Data

Heavy rain, earthquakes, and soil wetness are some of the natural causes of landslides inmountainous regions. Human actions, such as haphazard building, may also contribute to thisproblem. Effective forecast systems might lessen the impact of these catastrophes, which can causesevere property damage and loss of life. Recently, ml algorithms have been used for autonomouslandslip detection and prediction. The semi-automatic identification of landslides has made use ofsatellite data and a variety of feature extraction and classification methods. Still, there's a ways togo before completely automated detection can compete with human precision. One of the biggestchallenges is finding a trustworthy training database that produces accurate testing results. In orderto find where the current research is lacking, this study reviews all the methods that have beenemployed to date for landslip categorisation and detection using satellite imagery. It also suggests anew model for landslip forecasting. The accuracy and categorisation methods used in these articlesare examined, providing a window into the present and potential future. To further develop theapplication of ml, especially CNN models, for landslip detection from satellite images, thesuggested prototype employs dl models to enhance landslip detection and classification.