A Comprehensive Review of Flood Detection Innovations using IoT, GIS and Machine Learning

Abstract

Floods are among the most devastating natural disasters, necessitating timely and accurate detection to mitigate damage to infrastructure, ecosystems, and human life. Traditional early flood warning systems, while valuable, are increasingly limited by their inability to provide real-time data essential for live flood detection. Live detection relies on continuous monitoring of dynamic data streams, including IoT sensor networks, satellite imagery, and camera footage, enabling immediate response capabilities. This paper presents a systematic review of flood detection techniques, contrasting traditional methods with modern technologies such as IoT, Artificial Intelligence (AI), and Machine Learning (ML). We examine the role of IoT sensors in capturing real-time environmental data, as well as the integration of satellite imagery and live NVR (Network Video Recorder) footage. Furthermore, this review addresses challenges in accessing live data sources, as many real-time video feeds remain unavailable to the public, impacting detection effectiveness. Our findings offer insights into the advantages, limitations, and future directions for enhancing live flood detection systems.