MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641106659 A) filed by Kavyashree B; Dayananda Sagar Academy Of Technology And Management; Sushrut Bankalgi; Tejas R S; Patil Yash Laxman; and Mohammed Yahiya Khan on September 04, 2026, for Ai-Based Disaster Damage Assessment System.
Inventors include Kavyashree B; Dayananda Sagar Academy Of Technology And Management; Sushrut Bankalgi; Tejas R S; Patil Yash Laxman; and Mohammed Yahiya Khan.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: Recent advances in Artificial Intelligence (AI), computer vision, and Earth observation have significantly enhanced flood detection, inundation mapping, and infrastructure damage assessment. While conventional ground truth surveys remain constrained by safety risks, limited accessibility, and slow deployment, deep learning models applied to satellite, aerial, and unmanned aerial vehicle (UAV) imagery enable rapid and large-scale automated flood mapping. However, ensuring robust model generalization across diverse flood events, geographically varying terrain and building structures, and heterogeneous sensor resolutions remains a critical operational challenge. This review presents a systematic examination of state-of-the-art AI methodologies for flood assessment. We critically analyze the evolution of approaches spanning Convolutional Neural Networks (CNNs), U-Net-based semantic segmentation, Vision Transformers (ViTs), multi modal sensor fusion integrating SAR and optical imagery, and emerging self-supervised foundation models. Furthermore, we evaluate benchmark flood datasets, establish a transparent review methodology, identify key technical challenges such as domain shift, class imbalance, cloud interference, and edge deployment latency, and discuss promising research directions toward robust, scalable, and reliable AI-driven decision-support systems for flood response and emergency management.
Disclaimer: Curated by HT Syndication.
