DEEP LEARNING BASED CYCLONE INTENSITY ESTIMATION USING INSAT-3D IR IMAGERY

Artificial Intelligence

Classifying the severity of a given cyclone is one of the key aspects of cyclone forecasting. The risk to human lives and the harm the storm does to the environment can both be decreased by projecting the strength of the cyclone. The Dvorak approach has traditionally been used to estimate cyclone intensity. The technique's highlights on the analysis of the cyclone's cloud patterns poses one of its biggest difficulties. By automating the intensity estimation procedure and eliminating the distinction involved with manual analysis, CNNs and INSAT 3D images significantly contribute to resolving these difficulties. It is a diagnostic model since it can properly predict the intensity of tropical cyclones. Estimating the disaster's intensity is a crucial step for staying updated on it. The primary goal of this research project is to estimate cyclone intensity in order to prevent damage from cyclones, which can be quite dangerous. The identification of previously unrecognized patterns in the existence of cyclone intensity and irregularities in earlier observations. This might make it easier for people to understand how tropical cyclone intensity changes. We used satellite photography to find the tropical cyclone because of the ideology. The technology uses deep learning research with hurricane satellite data to provide an automated way for cyclone estimation. The model is fine-tuned to take into account a range of environmental factors, such as the state of the air and sea surface temperatures, that have an impact on cyclone strength. In the research, the deep learning model outperformed more traditional approaches in terms of cyclone intensity forecast accuracy, which is encouraging. The current system needs more time complexity for accurate evaluations of tropical cyclone strength.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.