YOLOv11-Based Human Victim Detection System from Real time Disaster Imagery

Natural disasters such as floods and earthquakes often result in severe loss of human life due to delays in victim identification and rescue operations.To address this challenge, this paper presents an automated victim detection and response system based on deep learning object detection techniques.The proposed system utilizes advanced YOLO-based models to detect victims from images and video streams captured through surveillance cameras and drones in disaster-affected areas.The input data is preprocessed and used to train a supervised detection model that accurately identifies victims in real time.The trained model is deployed for rapid detection, generating bounding boxes and confidence scores to locate victims efficiently.Upon detection, the system initiates response actions by visualizing victim locations on digital maps and storing detection data for further analysis.Experimental results demonstrate that the proposed approach achieves high detection accuracy and fast inference speed, making it suitable for real-time disaster management and rescue operations.The system significantly enhances situational awareness and supports timely decision-making for emergency response teams.