A Search Operation Unmanned Aerial Vehicle Using YOLO v5 with Real-Time Human Detection and Counting

Search and rescue operations are carried out in close collaboration with the community and as a team effort. When a community is affected by natural disasters the rescue operation takes place to save people that may be trapped by falling debris. The researchers developed an unmanned aerial vehicle with human detection and people counter using YOLOv5 for search operations. A YOLOv5 algorithm for object detection in recognizing humans in an image. In addition, the device has a people counter which counts the actual people in the frame and displays the total number of people discovered in real-time. The prototype was tested in different conditions and its reliability was computed. The calculated reliability of the system for the daytime is 100% while there is only 80% reliability for nighttime at 2500cm due to the lighting of the surroundings. Statistical analysis was also conducted to determine the accuracy of the prototype and obtained a p-value of less than 0.05 which means that there is a significant improvement in using YOLOv5 in detecting humans.