Extraction of SIFT Features for Identifying Disaster Hit areas using Machine Learning Techniques

Beyond the death toll, disasters have major and often devastating social and economic ramifications, including financial damage and loss. As a result of recent advances in artificial intelligence (AI), particularly in machine learning (ML) and deep learning (DL), disasters can be better managed. An emphasis has been placed on disaster and hazard prediction studies as well as assessments of risk and susceptibility and the detection of disasters as well as the implementation of early warning systems and disaster monitoring. A number of current disaster management applications have also been examined. The findings are discussed, and suggestions for further research are offered. As computing power rises and data availability expands, new flood detection methods are being developed using machine learning. To better understand nonlinear systems and make flood forecasts, machine learning has become a go-to tool. Hydrologic and hydraulic models that explain physical processes can be used in flood forecasting in conventional ways to predict hazard factors. Although these models may be used to gain a better understanding of a system, they frequently need a large amount of computing and data resources. Methods based on machine learning have the potential to enhance accuracy while also decreasing calculation time and model creation costs. As a result of this study, we now have a better understanding of the current status of flood forecasting using machine learning algorithms. This paper proposes a machine learning model for SIFT extraction. Once the extraction is done comparative analysis is done with the SVM and Fine Tree classifiers to prove the efficiency of the proposed model.