Published Paper Details:

Tropical cyclone intensity estimation is considered to play a vital role in disaster management, accurate and timely estimation of Cyclone intensity is crucial for disaster preparedness and mitigation. The existing system solely relies on a deep learning approach CNN where developing real-time cyclone intensity estimation systems can provide critical information for disaster response and evacuation planning. This requires efficient model architectures and hardware optimization to achieve real-time performance. This project aims at utilizing infrared imagery from INSAT 3D satellite with hybrid models for cyclone intensity estimation, CNN-RNN offer a promising approach to improve cyclone intensity estimation accuracy and address the limitations of individual CNN models. This comprehensive system aims in advanced cyclone forecasting, offering precise predictions of cyclone intensity and associated risks by synergistically incorporating spatial and temporal information.