For the current detection methods are not flexible and detection performance is not high, in this paper, we present a new method of video-based smoke detection algorithm by combining the improved codebook model and the Convolutional Neural Networks (CNNs). Firstly, the algorithm detects the suspected smoke regions by the improved codebook model. Secondly, it uses the deep Convolutional Neural Networks (CNNs) to extract the features of the suspected smoke area automatically, and then classify these features into smoke or non-smoke. Compared with the previous work, experimental results have shown that the detection precision in the testing sets can reach high performance. In addition, through the experiments on more than one video scene, it shows the effectiveness of our method and improves the smoke detection ability.

Date Published: 9 August 2018
PDF: 8 pages
Proc. SPIE 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018), 108065X (9 August 2018); doi: 10.1117/12.2502974

Published in SPIE Proceedings Vol. 10806:
Tenth International Conference on Digital Image Processing (ICDIP 2018)
Xudong Jiang; Jenq-Neng Hwang, Editor(s)