Open AccessComputer Vision based Early Electrical Fire-detection in Video Surveillance oriented for Building environment
Author(s) -
P. Sridhar,
R. R. Sathiya
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1916/1/012024
Subject(s) - fire detection , computer science , frame (networking) , frame rate , work (physics) , artificial intelligence , alarm , architecture , object detection , constant false alarm rate , deep learning , false alarm , computer vision , real time computing , architectural engineering , engineering , pattern recognition (psychology) , telecommunications , electrical engineering , mechanical engineering , art , visual arts
This work presents autonomous electrical fire-detection and localization using computer vision based techniques. The proposed work uses YOLO v2 to extract the electrical fire features more effectively than other conventional and machine learning approaches. This working model is tested on commercial and residential building as well as indoor and outdoor environments. This framework has achieved high detection accuracy and low false alarm rate. Besides, the proposed frame work can be used for early real-time electrical fire detection in surveillance videos and we present experimental results for electrical fire localization in CCTV footage using the deep learning architecture proposed in this work.