The Global flooding, exacerbated by shifting climate patterns and increased rainfall, poses a severe threat to lives and property. Swift rescue is essential, and unmanned aerial vehicles (UAVs) play a crucial role by providing real-time aerial perspectives. Manual operations are challenging, prompting a need for efficient data processing. Leveraging YOLO algorithm advancements in deep learning by swiftly and accurately identifying individuals in UAV images, improving rescue efforts during floods and similar disasters efforts, the disaster response has been enhanced. Utilizing YOLOv8 is pivotal for efficient and accurate disaster response. Its advanced object detection capabilities, especially in real-time scenarios, significantly enhance the speed and precision of identifying and locating individuals in emergency situations like floods. The model’s effectiveness in processing substantial datasets makes it a valuable asset in optimizing rescue efforts, ensuring a rapid and targeted response to areas impacted by calamities. In this paper assessment, tests have been conducted using datasets comprising 20, 40, and 60 images for a thorough evaluation. The generated confusion matrices enabled us to derive crucial performance metrics. Notably, the obtained accuracy stands at an impressive 89.30%, showcasing the model’s proficiency in accurately classifying instances across varying test image quantities. This strong performance, as reflected in precision, recall, and F1 score calculations, underscores the reliability of the proposed model in handling diverse datasets with consistently high precision and recall rates.