• DR. DINESHA L
• ASHA U NAYAK
• ASHWINI
• AVANTIKA AJAY PODUVAL
• BHAGYALAKSHMEE

One of the most destructive natural calamities, floods pose serious risks to both human life and property. Reducing the number of fatalities and financial losses requires prompt detection and rescue efforts. Through the detection and localization of live forms in submerged areas during floods, this research proposes a novel method that uses computer vision techniques to improve disaster response operations. In difficult flood situations, the suggested method accurately and efficiently identifies life forms by combining two cutting-edge deep learning models: YOLOv8 for object detection and U-Net for semantic segmentation. Segmentation and detection are the two primary parts of the strategy. In order to identify flooded areas and retrieve crucial spatial data for accurate analysis, the segmentation step uses U-Net. YOLOv8 is used in the detection phase to locate and identify live forms in the areas that have been segmented. The technology guarantees precise localization and contextual understanding by integrating the advantages of both models, which makes rescue operations more efficient. The system is a useful tool for emergency response teams because it is made to function in real-time. Through thorough preprocessing and model improvement, it also tackles important issues in disaster scenarios, like ambient noise and fluctuating illumination conditions. In addition to improving detection accuracy, the combination of segmentation and detection operations guarantees computational efficiency. By presenting an AI-powered approach that can greatly enhance rescue results during floods, this research aids in disaster management. In order to increase the system's scalability and accessibility in practical situations, future research will concentrate on adding multi-modal inputs, lightweight structures, and drone-based deployments.

Flood detection, Computer vision, Deep learning, U-Net, YOLOv8, Semantic segmentation, Object detection, Disaster management.

"FLOOD MONITORING AND LIFE DETECTION IN SUBMERGED AREAS USING COMPUTER VISION TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 12, page no.a194-a202, December-2024, Available :http://www.jetir.org/papers/JETIR2412020.pdf

"FLOOD MONITORING AND LIFE DETECTION IN SUBMERGED AREAS USING COMPUTER VISION TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 12, page no. ppa194-a202, December-2024, Available at : http://www.jetir.org/papers/JETIR2412020.pdf