Features selection for video smoke detection using random forest

WEN Zebo; KANG Yu; CAO Yang; WEI Meng; SONG Weiguo

doi:10.3969/j.issn.0253-2778.2017.08.004

JUSTC > 2017 > 47(8): 653-664. > DOI: 10.3969/j.issn.0253-2778.2017.08.004

PDF (6036 KB)

Open Access JUSTC Original Paper

Features selection for video smoke detection using random forest

1.
Department of Automation, University of Science and Technology of China, Hefei 230027, China
2.
State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei 230027, China

Cite this: JUSTC, 2017, 47(8): 653-664

https://doi.org/10.3969/j.issn.0253-2778.2017.08.004

Funds: Supported in part by National Natural Science Foundation of China (61422307, 61673361).

More Information

Author Bio:
WEN Zebo, male, born in 1992, Master candidate. Research field: Control theory. E-mail: dilong@mail.ustc.edu.cn
Corresponding author:
KANG Yu
Received Date: October 08, 2016
Revised Date: November 06, 2016
Published Date: August 30, 2017

Full text PDF

Abstract

Abstract

Using the random forest algorithm, a video smoke detection method with features selection was proposed. The method first extracted four original smoke image features including color features in RGB space, wavelet high frequency sub-images, multi-scale local max saturation, and multi-scale dark channel to input the random forest(RF). Then it utilized haze image formation model to make the synthetic smoke images from non-smoke images and partitions these images into blocks as the samples for RF. Thirdly, it trained RF to get the selected features from the original features and used support vector machine(SVM) to get a classifier which recognizes the smoke blocks and the non-smoke blocks. And then the smoke region candidate can be extracted from video images by the classifier. Finally, the method analyzed the detected smoke region with the features of the growth rate and the perimeter to area ratio to make the final decision on video smoke detection. The experimental results show that the proposed method can detect the smoke timely and give a fire alarm with a lower false-alarm rate.

Abstract

Using the random forest algorithm, a video smoke detection method with features selection was proposed. The method first extracted four original smoke image features including color features in RGB space, wavelet high frequency sub-images, multi-scale local max saturation, and multi-scale dark channel to input the random forest(RF). Then it utilized haze image formation model to make the synthetic smoke images from non-smoke images and partitions these images into blocks as the samples for RF. Thirdly, it trained RF to get the selected features from the original features and used support vector machine(SVM) to get a classifier which recognizes the smoke blocks and the non-smoke blocks. And then the smoke region candidate can be extracted from video images by the classifier. Finally, the method analyzed the detected smoke region with the features of the growth rate and the perimeter to area ratio to make the final decision on video smoke detection. The experimental results show that the proposed method can detect the smoke timely and give a fire alarm with a lower false-alarm rate.

FullText(HTML)

References (12)

References

[1]	CHEN T H, YIN Y H, HUANG S F, et al. The smoke detection for early fire-alarming system base on video processing[C]// Proceedings of the International Conference on Intelligent Information Hiding and Multimedia. Pasadena, USA: IEEE Computer Society, 2006: 427-430.
[2]	TREYIN B U, DEDEOGˇLU Y, ETIN A E. Wavelet based real-time smoke detection in video[C]// European Signal Processing Conference. Antalya, Turkey: IEEE Press, 2005: 1-4.
[3]	ALEJANDRO O B, LEONARDO M G, GABRIEL S P, et al. Improvement of a video smoke detection based on accumulative motion orientation model[C]// Electronics, Robotics and Automotive Mechanics Conference. Cuernavaca, Morelos, Mexico: IEEE Press, 2011: 126-130.
[4]	YUAN F. A fast accumulative motion orientation model based on integral image for video smoke detection[J]. Pattern Recognition Letters, 2008, 29(7): 925-932.
[5]	XU Z G, XU J L. Automatic fire smoke detection based on image visual features[C]// International Conference on Computational Intelligence and Security Workshops. Harbin, China: IEEE Press, 2007: 316-319.
[6]	YANG J, CHEN F, ZHANG W D. Visual-based smoke detection using support vector machine[C]//Proceedings of the 4th International Conference on Natural Computation. Jinan, China: IEEE Press, 2008,4: 301-305.
[7]	HORNG W B, PENG J W. Image-based fire detection using neural networks[C]// Proceedings of the 9th Joint International Conference on Information Sciences. Kaohsiung, Taiwan, China: Atlantis Press, 2006, [2016-10-01] http://xueshu.baidu.com/s?wd=paperuri%3A%2890d519bf26ad093adab4b2c913f00ada%29&filter=sc_long_sign&tn=SE_xueshusource_2kduw22v&sc_vurl=http%3A%2F%2Fdx.doi.org%2F10.2991%2Fjcis.2006.301&ie=utf-8&sc_us=1312074543806948226.
[8]	TUNG T X, KIM J M. An effective four-stage smoke-detection algorithm using video images for early fire-alarm systems[J]. Fire Safety Journal, 2011, 46(5): 276-282.
[9]	HE K M, SUN J, TANG X O. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353.
[10]	TANG K, YANG J, WANG J. Investigating haze-relevant features in a learning framework for image dehazing[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA: IEEE Press, 2014: 2995-3002.
[11]	CRIMINISI A, SHOTTON J, KONUKOGLU E. Decision forests: A unified framework for classification, regression, density estimation, manifold learning and semi-supervised learning[J]. Foundations and Trendsin Computer Graphics and Vision, 2012, 7(2-3): 81-227.
[12]	GENUER R, POGGI J M, TULEAU-MALOT C. Variable selection using random forests[J]. Pattern Recognition Letters, 2010, 31(14): 2225-2236.

Cited By

Track Citations

Get Citation

{{if article.articleBusiness.pdfLink && article.articleBusiness.pdfLink != ''}} {{else}} {{/if}}PDF

XML

References

[1]	CHEN T H, YIN Y H, HUANG S F, et al. The smoke detection for early fire-alarming system base on video processing[C]// Proceedings of the International Conference on Intelligent Information Hiding and Multimedia. Pasadena, USA: IEEE Computer Society, 2006: 427-430.
[2]	TREYIN B U, DEDEOGˇLU Y, ETIN A E. Wavelet based real-time smoke detection in video[C]// European Signal Processing Conference. Antalya, Turkey: IEEE Press, 2005: 1-4.
[3]	ALEJANDRO O B, LEONARDO M G, GABRIEL S P, et al. Improvement of a video smoke detection based on accumulative motion orientation model[C]// Electronics, Robotics and Automotive Mechanics Conference. Cuernavaca, Morelos, Mexico: IEEE Press, 2011: 126-130.
[4]	YUAN F. A fast accumulative motion orientation model based on integral image for video smoke detection[J]. Pattern Recognition Letters, 2008, 29(7): 925-932.
[5]	XU Z G, XU J L. Automatic fire smoke detection based on image visual features[C]// International Conference on Computational Intelligence and Security Workshops. Harbin, China: IEEE Press, 2007: 316-319.
[6]	YANG J, CHEN F, ZHANG W D. Visual-based smoke detection using support vector machine[C]//Proceedings of the 4th International Conference on Natural Computation. Jinan, China: IEEE Press, 2008,4: 301-305.
[7]	HORNG W B, PENG J W. Image-based fire detection using neural networks[C]// Proceedings of the 9th Joint International Conference on Information Sciences. Kaohsiung, Taiwan, China: Atlantis Press, 2006, [2016-10-01] http://xueshu.baidu.com/s?wd=paperuri%3A%2890d519bf26ad093adab4b2c913f00ada%29&filter=sc_long_sign&tn=SE_xueshusource_2kduw22v&sc_vurl=http%3A%2F%2Fdx.doi.org%2F10.2991%2Fjcis.2006.301&ie=utf-8&sc_us=1312074543806948226.
[8]	TUNG T X, KIM J M. An effective four-stage smoke-detection algorithm using video images for early fire-alarm systems[J]. Fire Safety Journal, 2011, 46(5): 276-282.
[9]	HE K M, SUN J, TANG X O. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353.
[10]	TANG K, YANG J, WANG J. Investigating haze-relevant features in a learning framework for image dehazing[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA: IEEE Press, 2014: 2995-3002.
[11]	CRIMINISI A, SHOTTON J, KONUKOGLU E. Decision forests: A unified framework for classification, regression, density estimation, manifold learning and semi-supervised learning[J]. Foundations and Trendsin Computer Graphics and Vision, 2012, 7(2-3): 81-227.
[12]	GENUER R, POGGI J M, TULEAU-MALOT C. Variable selection using random forests[J]. Pattern Recognition Letters, 2010, 31(14): 2225-2236.

[1]	Haotian Wu, Bingqian Lin, Canjun Li, Wenping Zeng, Lili Qu, Chunlei Cang. Sphingosine-1-phosphate induces Ca²⁺ mobilization via TRPC6 channels in SH-SY5Y cells and hippocampal neurons[J]. JUSTC, 2022, 52(9): 2-1-2-10. DOI: 10.52396/JUSTC-2022-0014
[2]	GU Ruizhe, WANG Ping, ZHENG Mengce, HU Honggang, YU Nenghai. Adversarial attack based countermeasures against deep learning side-channel attacks[J]. JUSTC, 2020, 50(10): 1343. DOI: 10.3969/j.issn.0253-2778.2020.10.006
[3]	WANG Xin, ZHU Yangguang, WU Zun. Cooperation mode selection and pricing between hotel and OTA considering consumer preference under dual-channel supply chain[J]. JUSTC, 2020, 50(7): 985-992. DOI: 10.3969/j.issn.0253-2778.2020.07.016
[4]	LUSIANI A. (for the BaBar Collaboration). Search for dark photon and long-lived particles at BaBar[J]. JUSTC, 2016, 46(7): 601-607. DOI: 10.3969/j.issn.0253-2778.2016.07.010
[5]	GONNELLA Francesco (for the NA/ Collaboration). Search for dark photons in π0 decays at NA48 and NA62[J]. JUSTC, 2016, 46(7): 594-600. DOI: 10.3969/j.issn.0253-2778.2016.07.009
[6]	PREZ DEL RO Elena (for the KLOE- Collaboration). Dark mediator searches at KLOE/KLOE-2[J]. JUSTC, 2016, 46(7): 587-593. DOI: 10.3969/j.issn.0253-2778.2016.07.008
[7]	JIANG Bin, WANG Hai-ming, GAO Xi-qi, YOU Xiao-hu. Two-dimensional DCT-based channel estimation for OFDMA systems[J]. JUSTC, 2009, 39(10): 1114.
[8]	JI Jie, HUO Ran, ZHANG Ying, ZHONG Wei, HU Long-hua. Experimental study on the entrainment mass flow rate across the smoke layer interface during horizontal spread in a long channel[J]. JUSTC, 2009, 39(7): 738-742.
[9]	JIANG Ya-qiang, HUO Ran, HU Long-hua, JI Jie, YANG Dong, WANG Hao-bo. A numerical study on smoke layering plugholing under mechanical extraction in a long channel[J]. JUSTC, 2009, 39(4): 424-428.
[10]	XU Kan, LIU Ming-hou, LIU Dong, WANG Ya-qing. Study on mini-channel structured radiator filled with porous media[J]. JUSTC, 2009, 39(1): 63-68.

TrendMD

Volume 47 Issue 8 PP. 653-664

Cover

Keywords

Article Metrics

Article views (615) PDF downloads (295)

Features selection for video smoke detection using random forest

Abstract

Abstract

References

Related Articles

Catalog

References

Related Articles

TrendMD

Article Metrics

Authors

Browse

Contact Us

About

Features selection for video smoke detection using random forest

Share

Tools

Abstract

Abstract

References

Related Articles

Catalog

References

Related Articles

TrendMD

Article Metrics

Authors

Browse

Contact Us

About

Export File

Citation

Format

Content