Cyclone Intensity Prediction Using Deep Learning on INSAT-3D IR Imagery: A Comparative Analysis

Cyclone Intensity Prediction Using Deep Learning on INSAT-3D IR Imagery: A Comparative Analysis

Ayush Kiranpure1, Bhavesh Netankar2, Rushikesh Wankhade3, Yash Asalmol4, Mihir Raichura5, Prof. Y. A. Thakare6

1,2,3,4,5Undergraduate Student, Sipna College of Engineering and Technology, Amravati (MH), India
6Assistant Professor, Sipna College of Engineering and Technology, Amravati (MH), India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract:

This study investigates the effectiveness of deep learning techniques in accurately estimating tropical cyclone intensity using infrared (IR) imagery from the INSAT-3D satellite. We assess the performance of three models—Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and a hybrid CNN-RNN model—comparing them against traditional machine learning methods like Support Vector Machines (SVM) and Random Forests (RF). Results demonstrate that deep learning models significantly outperform traditional approaches, with the CNN-RNN model achieving the highest accuracy. These findings highlight the potential of deep learning to enhance early warning systems for extreme weather events.

Keywords: Deep learning, Machine Learning, Preprocessing , CNN, INSAT 3D Images