Novel Deep Clustering Network Technique for Passive Microwave Rainfall Infrared Images Assessment of Tropical Cyclones

The energy that tropical cyclones (TCs) acquire from ocean water vapor makes them one of the most destructive forms of weather. The substantial water vapor content carried by these cyclones often results in heavy rainfall and severe flooding. Timely estimation of TC rainfall is crucial for disaster warning and mitigation efforts. However, accurately estimating passive microwave (PMW) rainfall from tropical cyclones, with high geographical and temporal resolution, remains challenging due to the limited spatial range of microwave sensors. In this research, we propose an innovative deep clustering network algorithm to assess TC behaviour using infrared pictures of PMW. The experimental findings demonstrate that our proposed approach surpasses existing models, hence presenting novel opportunities for enhanced understanding of TC and improved prediction of disasters.