<![CDATA[Fire detection is one of the technological efforts to prevent fire incidents. This is very important because the damage caused by fires can be minimized by having a fire detector. There are two types of fire detection, namely traditional-based and computer vision-based. Traditional-based fire detection has many shortcomings, one of which requires a close fire distance for activation. Hence, computer vision-based fire detection is made to cover the shortcomings of traditional-based fire detection. Therefore, in this study, we propose a video-based fire detection using a Convolutional Neural Network (CNN) Deep Learning approach supported by You Only Look Once (YOLO) object detection model version four. This study uses a dataset of various fire scenarios in the form of images and videos. The fire detection built in this study has an accuracy of above 90% with an average detection speed of 34.17 Frame Per Second (FPS).]]>
Copyrights © 2022
