| Title |
Detecting Location of Fire in Video Stream Environment using Deep Learning |
| Authors |
김윤지(Yun-ji Kim) ; 조현종(Hyun-chong Cho) |
| DOI |
https://doi.org/10.5370/KIEE.2020.69.3.474 |
| Keywords |
CCTV Video Analysis; Convolution Neural Network; Deep Learning; Fire Detection |
| Abstract |
To avoid the large scale of damage of fire occurred, it is necessary to have a system to detect the incident as soon as possible.
Traditional sensors and vision based systems for fire detection is limited in indoors and need more computational time and memory, restricting its implementation. In this paper, we propose a video-based fire detection system using deep learning to solve these problems. To run real-time detection in video stream, the activity detection is performed within a single image frame and makes prediction with a single network evaluation. The object detection algorithm we applied can tell the location as well as the presence of fire. It allows us to analyze the cause of the fire through video and monitor extensive areas efficiently. The results of the proposed system showed 99% precision, 99% accuracy and 100% recall. Experimental results show that the proposed method has excellent fire detection performance. |