Detection of Natural Disaster Victims using You Only Look Once (YOLO)
Moechammad Sarosa, Nailul Muna, Erfan Rohadi

State Polytechnic of Malang

Abstract

Natural disasters are events that cannot be predicted both by location and time of occurrence. Natural disasters cause property losses and can even take lives. The handling of rapid evacuation must be done by the SAR team to help victims of natural disasters to reduce the amount of loss. But in reality, there are many obstacles in the evacuation process. Starting from the difficulty of the terrain that is reached to the limited equipment needed. In this research, a system designed to detect victims of natural disasters uses image processing where the shooting of victims is carried out using a drone that aims to help find victims in difficult or vulnerable locations when reached directly by humans. Against this background, this research proposes the development of a method for the detection of victims of natural disasters that aims to assist the SAR team and natural disaster volunteers in searching for victims who are in hard to reach places. The method used is You Only Look Once (YOLO) using the python programming language related to image processing. From the research that has been done, the accuracy of detecting objects of disaster victims has good results

Keywords: Image Processing, Natural Disasters, Object Detection, Victim Detection, You Only Look Once (YOLO)

Topic: Computer Science