Paper Deep Learning - An Investigation on Fire and Fire's Losses Detection Using YOLOv5

the safety of people's lives and their properties is greatly threatened by fire, a common and highly disastrous incident. Consequently, fire and surrounding tools detection are significant for safety examination, and an early warning may help people efficiently decrease losses. Utilizing temperature sensors, smoke detectors, thermal cameras, and other traditional fire detection techniques has apparent flaws. These tools are frequently expensive, weak, and not accurate. Humans have recently been a part of a highly informational age. Specifically, the rapid progress in deep learning opens up new avenues for fire and surrounding tools and fire detection. The data transfer rate issue is resolved, but the terminal's detecting capabilities still need to be improved. Computer vision-based fire detection technology has started reaching the public's vision field. This paper mainly focuses on the benefits of YOLOv5 and its uses in detecting fire and related losses. Moreover, the researcher runs experimentation using the most recent YOLOv5 target identification algorithm to get to the findings. According to the experimental findings, YOLOv5-based fire and fire loss detection can achieve higher accuracy. A fire and fire loss detection system studied in this paper would significantly impact the security field, particularly physical security.