Abstract
Forest fire is an serious hazard in many places around the world. For such threats, video-based smoke detection would be particularly important for early warning because smoke arises in any forest fire and can be seen from a long distance. This paper presents a novel and robust approach for smoke detection that employs Deep Belief Networks. The proposed method is divided into three phases. In the preprocessing phase, the region of high motion is extracted by background subtraction method. During the next phase, smoke pixel intensities are extracted from the Red, Green and Blue and Luminance; Chroma:Blue; Chroma:Red color spaces for foreground regions. Subsequently, second feature which is based on texture is computed for detecting smoke regions in which Local Extrema Co-occurrence Pattern, an improved version of local binary patterns are extracted from different foreground regions which compute not only texture of smoke but also intensity and color of smoke using Hue Saturation Value color space. Finally, Deep Belief Network is employed for classification. The proposed method proves its accuracy and robustness when tested on different varieties of scenarios whether wildfire-smoke video, hill base smoke video, indoor or outdoor smoke videos.
Similar content being viewed by others
Notes
http://cvpr.kmu.ac.kr/.
http://www.openvisor.org.
http://signal.ee.bilkent.edu.tr/VisiFire/Demo.
https://www.shutterstock.com/video/search/smoke.
https://sites.google.com/site/smokedataset/smokedataset.
References
Amiaz T, Fazekas S, Chetverikov D, Kiryati N (2007) Detecting regions of dynamic texture. In: International conference on scale space and variational methods in computer vision, pp 848–859. Springer
Calderara S, Piccinini P, Cucchiara R (2008) Smoke detection in video surveillance: a mog model in the wavelet domain. In: International conference on computer vision systems, pp. 119–128. Springer
Celik T, Demirel H, Ozkaramanli H, Uyguroglu M (2007) Fire detection using statistical color model in video sequences. J Vis Commun Image Represent 18(2):176–185. doi:10.1016/j.jvcir.2006.12.003
Çelik T, Ozkaramanli H, Demirel H (2007) Fire and smoke detection without sensors: image processing based approach. In: Signal processing conference, 15th European. Citeseer
Chen J, He Y, Wang J (2010) Multi-feature fusion based fast video flame detection. Build Environ 45(5):1113–1122. doi:10.1016/j.buildenv.2009.10.017
Chetverikov D, Péteri R (2005) A brief survey of dynamic texture description and recognition. In: Computer recognition systems pp 17–26. Springer
Cui Y, Dong H, Zhou E (2008) An early fire detection method based on smoke texture analysis and discrimination. In: Image and signal processing, congress on, vol 3, pp 95–99. IEEE
Doretto G, Chiuso A, Wu YN, Soatto S (2003) Dynamic textures. Int J Comput Vis 51(2):91–109 . doi:10.1023/A:1021669406132
Enis Cetin A, Porikli F (2011) Special issue on dynamic textures in video. Mach Vis Appl 22(5):739–740. doi:10.1007/s00138-011-0325-0
Fazekas S, Amiaz T, Chetverikov D, Kiryati N (2009) Dynamic texture detection based on motion analysis. Int J Comput Vis 82(1):48–63. doi:10.1007/s11263-008-0184-y
Günay O, Taşdemir K, Töreyin BU, Çetin AE (2010) Fire detection in video using lms based active learning. Fire Technol 46(3):551–577. doi:10.1007/s10694-009-0106-8
Habiboglu YH, Gunay O, Cetin AE (2011) Real-time wildfire detection using correlation descriptors. In: Signal processing conference, 19th European, pp 894–898. IEEE
Hinton GE, Osindero S, Teh YW (2006) A fast learning algorithm for deep belief nets. Neural Comput 18(7):1527–1554
Hinton GE, Salakhutdinov RR (2006) Reducing the dimensionality of data with neural networks. Science 313(5786):504–507
Kopilovic I, Vagvolgyi B, Szirányi T (2000) Application of panoramic annular lens for motion analysis tasks: surveillance and smoke detection. In: Conference on pattern recognition, 15th international, vol 4, pp 714–717. IEEE
Liu ZG, Yang Y, Ji XH (2016) Flame detection algorithm based on a saliency detection technique and the uniform local binary pattern in the ycbcr color space. Signal Image Video Process 10(2):277–284. doi10.1007/s11760-014-0738-0
Palm RB (2012) Prediction as a candidate for learning deep hierarchical models of data. Technical University of Denmark, p 5
Phillips Iii W, Shah M, da Vitoria Lobo N (2002) Flame recognition in video. Pattern Recognit Lett 23(1):319–327 (2002). doi:10.1016/S0167-8655(01)00135-0
Piccinini P, Calderara S, Cucchiara R (2008) Reliable smoke detection in the domains of image energy and color. In: Conference on image processing, 15th international, pp 1376–1379. IEEE
Qi X, Ebert J (2009) A computer vision based method for fire detection in color videos. Int J Imag 2(S09):22–34
Verma M, Raman B, Murala S (2015) Local extrema co-occurrence pattern for color and texture image retrieval. Neurocomputing 165:255–269
Vicente J, Guillemant P (2002) An image processing technique for automatically detecting forest fire. Int J Thermal Sci 41(12):1113–1120. doi:10.1016/S1290-0729(02)01397-2
Xu Z, Xu J (2007) Automatic fire smoke detection based on image visual features. In: International conference on computational intelligence and security workshops, CISW, pp 316–319. IEEE
Ye W, Zhao J, Wang S, Wang Y, Zhang D, Yuan Z (2015) Dynamic texture based smoke detection using surfacelet transform and hmt model. Fire Saf J 73:91–101
Yuan F (2011) Video-based smoke detection with histogram sequence of lbp and lbpv pyramids. Fire Saf J 46(3):132–139. doi:10.1016/j.firesaf.2011.01.001
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Pundir, A.S., Raman, B. Deep Belief Network For Smoke Detection. Fire Technol 53, 1943–1960 (2017). https://doi.org/10.1007/s10694-017-0665-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10694-017-0665-z