Research on Detection Techniques of Early Forest Fire Based on Dynamic Characteristics of Smoke

Abstract:

The forest fire has been threatening the forest ecosystem and has brought huge economic losses to humans. Traditional fire detection systems use ion-optical smoking type and other physical or chemical means to discover a fire, which is not suitable for outdoor forest fires with long-distance and large-area characteristics. This paper presents the early fire detection algorithm based on smoke dynamic characteristics and its main part includes smoke color model, dynamic feature extraction and fire area connected component analysis. Through standard video database performance testing and compared to the algorithm and the traditional fire detection method, we can achieve early warning of a fire and can effectively reduce false and negative rate of fire monitoring system.

• Periodical: Applied Mechanics and Materials (Volume 340)
• Pages: 431-435
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.340.431
• Online since: July 2013
• Authors: Hong Lin, Zhen Ping Qiang, Guang Zhi Di, Zhi Gang Liu
• Keywords: Connected Domain Analysis, Dynamic Characteristic, Forest Fires, Smoke Color Model

References

[1] F Liu, X Wang. (2009) Digital video image processing and communication . Mechanical industry press, (12): 79-81.

[2] L Wu, F Peng, K Zhao, (2012) analysis of the forest fire smoke real-time detection method based on characteristics of the movement. Signal processing, (9): 55-57.

[3] Y Li. (2010) Digital visual video technology . Xi'an electronic university of science and technology press, 57-60.

[4] L Wu, J Fang. (2010) Fire detection and information processing . Chemical industry press, 511-514.

[5] Y Zhang. (2010)Image understanding and computer vision. Beijing: Tsinghua university press, 778-782.

[6] T Chen, Y Yin, Sh Huang and Y Ye. (2006).

[7] T Chen, Y Yin, Sh Huang and Y Ye, The Smoke Detection for Early Fire-Alarming System Base on Video" Proceedings of the 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP, 06) Processing,: 2078-(2084).

DOI: 10.1109/iih-msp.2006.265033

[8] TChen, Y Yin, Sh Huang and Y Ye, (2004) An Early Fire-Detection Method Based on Image Processing, International Conference on Image Processing (ICIP): 1325-1332.

[9] T.H. Chen, C.L. Kao and S.M. Chang , (2003) An Intelligent Real-time Fire Detection Method Based on Video Processing , in Processing of the IEEE 37th Annual (2003) International Carnahan Conference on Security Technology., pp , 104-111.

DOI: 10.1109/ccst.2003.1297544

[10] T. Celik et al., Apr 2007 Fire Detection Using Statistical Color Model in Video Sequences, J. Visual Commun. Image Representation, vol. 18, no. 2, pp.176-185.

DOI: 10.1016/j.jvcir.2006.12.003

[11] L Yang , (2002) Fire Situation and Fire characteristic analyses is based on fire statistics of China, Fire safety journal, 21: 202-209.

DOI: 10.1016/s0379-7112(01)00054-6

[12] T. Celik, H. Demirel, and H. Ozkaramanli, (2006) Automatic Fire Detection in Video Sequences, Proc. European Signal Process. Conf., Florence, Italy, Sept. 10(9): 117-121.

[13] A Xu, L Fang , X Lou . Mar. (2010) Forest fire detection algorithm based on visible light video, Beijing Forestry University, Vol 32, No. 2, 02(11)33-39.

[14] L Wu, F Peng, K Zhao. Real-time forest fire smoke detection method based on motion feature analysis, signal processing, 93-97.

[15] Y Li. Apr, (2010) . Digital visual video technology, Xi'an Electronic Science and Technology University Press, 2006. V01. 22. No. 2, 98-104.

[16] SH X, M Zhao, J Xu . Jun, (2008) The field of early fire image recognition method, Computer Technology and Development, Vol 18 No. 62(4): 38-45.