Open AccessDetecting body parts from natural disaster victims using You Only Look Once (YOLO)
Author(s) -
S Moechammad,
Reiza Adi Cahya,
A N A Berkah
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1073/1/012062
Subject(s) - natural disaster , computer science , terrain , python (programming language) , drone , natural (archaeology) , artificial intelligence , computer security , geography , cartography , archaeology , biology , meteorology , genetics , operating system
Natural disasters are unpredictable events in both the location and the time of occurrence. Natural disasters can cause property loss and can even be claimed by life. To reduce the amount of losses, the handling of rapid evacuation should be conducted by the SAR team to help victims of natural disasters. But in fact, there are a lot of obstacles in the evacuation process. Starting from the difficulty of searching the victim’s body, the difficulty of the terrain reached until limited equipment needed. In this study designed the body detection system of natural disaster victims using image processing where the shooting of victims was carried out using drones aiming to help find victims in a difficult or prone location when reached directly by humans. Background of the problem, in this research proposed a development method for the detection of victims of natural disaster that aims to help the SAR team as well as natural disaster volunteers in the search for victims who are in a difficult to reach place. The method used by You Only Look Once (YOLO) uses the Python programming language associated with image processing. From the research has been obtained accuracy detection object disaster victims with good accuracy. Based on the experiments that have been done obtained a good accuracy value of 95.49% with epoch of 12000.