Abstract
Tropical cyclones are big storms that can cause a lot of damage. Cyclone intensity estimation plays a vital role in disaster preparedness, response, and mitigation strategies. This paper introduces a novel approach to estimating cyclone intensity using satellite images through a Convolutional Neural Network (CNN) model. Unlike previous methods, we employ advanced techniques such as histogram analysis for feature extraction and adaptive thresholding for image segmentation using mean, Gaussian, and Otsu methods. The model also predicts potential coverage distance. Additionally, we present a user-friendly visualization portal, a pioneering effort in this field, which displays the deep learning output along with contextual information for end-users
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call