基于运动特征的林火烟雾图像检测技术研究
|
摘要
森林是人类生存与发展不可或缺的资源,林火是造成森林资源破坏的最主要灾害。因此及时发现火情,对尽早扑灭林火减少森林资源损失有着重要的意义。目前,我国大部分林区虽然都建立了视频林火监控网络,但火情判断主要还是依靠人工观测监控。在现有的硬件系统基础上加入林火智能识别算法,不仅能够实现火情自动报警,而且改造成本较低。烟雾是森林火灾前期最为明显的特征,所以基于可见光视频的烟雾智能识别研究具有更加重要的意义。
本文通过理论仿真分析与工程试验,提出了一个森林火灾烟雾识别算法框架。首先根据烟雾的运动特性,利用运动检测算法确定烟雾疑似区域,再对疑似区域提取图像特征,使用训练好的分类器进行智能识别排除干扰物影响。
主要成果如下：
(1)提出灰度投影算法对林火视频序列图像进行电子稳像,减小了大风造成的摄像机抖动影响。
(2)引入运动检测法中的高斯模型背景减除法,利用火灾烟雾持续缓慢的扩散运动特性进行烟雾疑似区域提取,再使用带阈值的最小距离聚类法对零散琐碎区域聚合整理。
(3)基于烟雾较强的视觉特征,提取10维烟雾图像特征,利用Adaboost算法通过对训练样本集合的不同组合训练出多个弱分类器——分类与回归决策树。最终通过对弱分类器的组合达到强分类器的效果,解决了烟雾识别非线性可分问题。仿真实验达到漏报率4.53%、误报率5.91%。
(4)设计开发了森林火灾智能识别软件系统。在视频监控平台上融合了基于本文研究的序列图像去抖动、运动目标检测、图像特征提取以及Adaboost算法分类识别方法等,开发了森林火灾智能识别系统。经工程应用检验,系统运行稳定,识别准确率99.32%、误报率0.67%并成功捕获一次真实火情。
Forest is an important resource which is essential to human being's survival and development. As a major nature disaster, the forest fire damages large quantity of forest resources very year. In order to put out the forest fire as fast as possible and reduce the loss, it makes sense to detect the fire in time. Nowadays, video network has been established in most forests to monitor fire. However, it purely relies on artificial recognition. Forest fire video monitoring network based on hardware and advanced detection algorithm can not only automatically alarm of fire, but also decrease the cost significantly. Smoke is the most significant phenomenon prior to a fire, so research on video-based smoke recognition makes a good sense.
In this paper, the author proposed a frame of forest fire smoke recognition system. Several major results of this study are as follows:
(1) Gray projection algorithm which is one of electronic image stabilization methods is put forward to maintain the video series smooth and stable. The algorithm has good effect on image sequence with colorful and small timing jitter.
(2) Moving target detection algorithm based on Gaussian statistical model is proposed. Scattered trivial regions are clustered by minimum distance clustering with threshold.
(3)10dimension image features of forest fire smoke are extracted based on distinguishing visual characteristic of smoke. Multiple weak classifiers, classification and regression tree, is constructed with Adaboost algorithm. Strong classifier is established combined with those weak classifiers, and could be used to detect forest fire smoke. Simulation results show that the identification ratio is above94.7%, while false positive ratio was4.53%, false negative ratio was5.91%.
(4) Integration of wireless networks, video monitoring and other techniques contributes to the intelligent forest fire smoke recognition system. This system implements the results of this study, including electronic image stabilization, forest fire video motion detection algorithm, image feature extraction method and Adaboost classification and recognition method. It was deployed in a forest in Beijing, and successfully captured a true forest fire and all simulated forest fires tests. The identification ratio was99.32%.
本文通过理论仿真分析与工程试验,提出了一个森林火灾烟雾识别算法框架。首先根据烟雾的运动特性,利用运动检测算法确定烟雾疑似区域,再对疑似区域提取图像特征,使用训练好的分类器进行智能识别排除干扰物影响。
主要成果如下：
(1)提出灰度投影算法对林火视频序列图像进行电子稳像,减小了大风造成的摄像机抖动影响。
(2)引入运动检测法中的高斯模型背景减除法,利用火灾烟雾持续缓慢的扩散运动特性进行烟雾疑似区域提取,再使用带阈值的最小距离聚类法对零散琐碎区域聚合整理。
(3)基于烟雾较强的视觉特征,提取10维烟雾图像特征,利用Adaboost算法通过对训练样本集合的不同组合训练出多个弱分类器——分类与回归决策树。最终通过对弱分类器的组合达到强分类器的效果,解决了烟雾识别非线性可分问题。仿真实验达到漏报率4.53%、误报率5.91%。
(4)设计开发了森林火灾智能识别软件系统。在视频监控平台上融合了基于本文研究的序列图像去抖动、运动目标检测、图像特征提取以及Adaboost算法分类识别方法等,开发了森林火灾智能识别系统。经工程应用检验,系统运行稳定,识别准确率99.32%、误报率0.67%并成功捕获一次真实火情。
Forest is an important resource which is essential to human being's survival and development. As a major nature disaster, the forest fire damages large quantity of forest resources very year. In order to put out the forest fire as fast as possible and reduce the loss, it makes sense to detect the fire in time. Nowadays, video network has been established in most forests to monitor fire. However, it purely relies on artificial recognition. Forest fire video monitoring network based on hardware and advanced detection algorithm can not only automatically alarm of fire, but also decrease the cost significantly. Smoke is the most significant phenomenon prior to a fire, so research on video-based smoke recognition makes a good sense.
In this paper, the author proposed a frame of forest fire smoke recognition system. Several major results of this study are as follows:
(1) Gray projection algorithm which is one of electronic image stabilization methods is put forward to maintain the video series smooth and stable. The algorithm has good effect on image sequence with colorful and small timing jitter.
(2) Moving target detection algorithm based on Gaussian statistical model is proposed. Scattered trivial regions are clustered by minimum distance clustering with threshold.
(3)10dimension image features of forest fire smoke are extracted based on distinguishing visual characteristic of smoke. Multiple weak classifiers, classification and regression tree, is constructed with Adaboost algorithm. Strong classifier is established combined with those weak classifiers, and could be used to detect forest fire smoke. Simulation results show that the identification ratio is above94.7%, while false positive ratio was4.53%, false negative ratio was5.91%.
(4) Integration of wireless networks, video monitoring and other techniques contributes to the intelligent forest fire smoke recognition system. This system implements the results of this study, including electronic image stabilization, forest fire video motion detection algorithm, image feature extraction method and Adaboost classification and recognition method. It was deployed in a forest in Beijing, and successfully captured a true forest fire and all simulated forest fires tests. The identification ratio was99.32%.
引文
[1]边肇祺,张学工.模式识别(第二版)[M].北京：清华大学出版社,1999：176-211.
[2]程娟.复杂背景下运动目标识别算法研究[D].武汉理工大学,2008.
[3]戴南.基于决策树的分类方法研究而[D].南京师范大学,2003,1：4-26.
[4]冈萨雷斯.数字图像处理(第二版)[M].北京：电子工业出版社,2007：72-74.
[5]高爱莲.基于视频的火焰检测方法研究[D].昆明理工大学,2008.
[6]郭富强,于波,汪叔华.陀螺稳定装置及其应用[M].西安：西北工业大学出版社,1995.
[7]姜超.视频序列中运动目标提取与跟踪技术的研究[D].大连理工大学,2009.
[8]刘泉志.基于视频的运动目标检测与跟踪算法研究[D].上海交通大学,2009.
[9]刘听雨.基于视频特征的火灾火焰识别算法[D].西安电子科技大学.2009.
[10]罗鹏.林火协同监测技术研究[D].中国林业科学研究院,2010.
[11]罗琴娟.森林火灾视频图像智能识别技术研究[D].北京林业大学,2009.
[12]吕普铁.基于普通CCD摄像机的火灾探测技术研究[J].哈尔滨工程大学学报,2003,2.44-48
[13]康锋.基于视觉特征的早期农林火灾检测方法的基础研究[D].浙江大学,2010：31-33.
[14]李彦斌,陈树越.早期森林火灾烟雾图像序列的分割算法[J].计算机工程与应用,2008,44(11)：203-204.
[15]练秋生,李党.融合多种特征的烟雾图像检测算法[J].光学技术,2009,35(4)：523-528.
[16]Richard O.Duda,Peter E.Hart,David G Stork.模式分类[M].北京:机械工业出版社,2003: 356-412.
[17]任海,高昌海,顾香凤.欧美森林火灾监测新技术及设备[J].林业机械与木工设备,2005,33(5)：8-11.
[18]任厚平,张永明.基于混合高斯模型定位的火灾烟雾纹理特征提取[J].微计算机信息,2005(21).
[19]饶裕平,方陆明,等.林火视频监控中烟识别方法概述[J].四川林业科技,2009,30(1)：81-85.
[20]宋广玲.一种基于CART决策树改进算法[J].哈尔滨理工大学学报,2009,2.
[21]吴雪琼,覃先林,等.我国林火监测体系现状分析[J].内蒙古林业调查设计,2010,33(3)：69-72.
[22]王涛,刘渊,等.一种基于飘动性分析的视频烟雾检测新方法[J].电子与信息学报,2011,33(5)：1024-1029.
[23]西奥多里蒂斯等.模式识别(中文第二版)[M].北京：电子工业出版社,2004：106-131.
[24]肖刚.国内外森林防火技术现状及趋势探讨[D].天津大学,2005,(11)：3-65.
[25]徐鹏,任波.基于单高斯模型的森林火灾烟图像目标检测[J].计算机与现代化,2009,(2)：18-20.
[26]闫厚,倪金生等.基于网络摄像机的森林火灾烟火自动识别技术探讨[J].森林防火,2007(2)：25-28.
[27]姚志芳.加强森林防火教育的重要性及措施[J].现代农业科技,2008(14)：351-353.
[28]袁非牛.基于运动累积和半透明的视频烟雾探测模型[J].数据采集与处,2007,22(4)：396-399.
[29]郑焕能.森林防火[M].哈尔滨：东北林业大学出版社,1994.
[30]钟平.机载电子稳像技术研究[D].中国科学院长春光机所.2003.
[31]朱娟娟.电子稳像理论及其应用研究[D].西安电子科技大学,2009：3-7.
[32]Bigley W J.Optimal motion stabilization control of an electro-optiacl sight system[C].Proc.SPIE, 1989.Ⅳ:116-120.
[33]B.K.P.Hom and B.Schunk. Determining Optical Flow, Artificial Intelligence, Vol.17,1981: 185-204.
[34]B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin, "Flame detection in video using hidden markov models," ICIP'05, pp.1230-1233 [2005].
[35]B. U. Toreyin, R. G. Cinbis. Fire detection in infrared video using wavelet analysis, Optical Engineering 46(6).
[36]B.Vagvolgyi,I.Kopilovic and T.Sziranyi, Application of panoramic annular lens for motion analysis tasks:surveillance and smoke detection[C], Proceedings of 15th International Conference on Pattern Recognition,2000, Volume 4,3-7 Sept.2000:714-717.
[37]Damir Krstinic, Darko Stipanicev, Toni Jakovcevic. Histogram-based smoke segmentation in forest fire detection system[J]. Information Technology And Control,2009,38(3):237-244.
[38]F. Vella, A. Castorina, M. Mancuso, G. Messina. Digital image stabilization by adaptive block motion vectors filtering[J]. IEEE Transactions on ConsumerElectronics,2002,48(3):796-801.
[39]Gomez-Rodriguez F, Arrue B C, Ollero A. Smoke monitoring and measurement using image processing[C]. Application to forest fires. In. Proc. of SPIE, Automatic Target Recognition XIII, 2003,5094:404-411.
[40]H. J. Grech-Cini. Smoke Detection[J]. US Patent No. US6844818B2.
[41]Ham SunJae, Ko Byoung-Chul, Nam Jae-Yea. Vision based forest smoke detection using analyzing of temporal patterns of smoke and their probability models [C]. Proceedings of SPIE-The International Society for Optical Engineering,2011,78-77.
[42]Huang J C. Automatic feature-based global motion estimation in video sequences. IEEE,2004, 50(3):911-915.
[43]K.Sauer and B.Schwartz. Efficient block motion estimation using integral projections. IEEE Trans. on Circuits and Systems for Video Technology,1996,6(5):513-518
[44]Morimoto C, Chellappa R. Fast electronic digital image stabilization for off-road navigation. IEEE Journal of Real-time Imaging,1996(2):285-296.
[45]Myrheim J, Rue H. New algorithms for maximum entropy image restoration[J]. CVFIP,1992,54(3): 223-238.
[46]Ohta Y, Kandae T, Sakai T. Color information for region segmentation. Computer Graphics Image Process,13:pp.222-241,1980.
[47]Paulo Vinicius Koerich Borges. Efficient visual fire detection applied for video retrieval.
[48]R M Harliek, D Shnagmugna. Textural Feature for Image Classification, IEEE Transaction on Systems, Mna, Cybenreties, Vol.3(6), PP.610-621,1973.
[49]Surit S, Chatwiriya W. Forest fire smoke detection in video based on digital image processing approach with static and dynamic characteristic analysis [C].2011 First ACIS/JNU International Conference on CNSI Proceedings,2011:35-39.
[50]Soliman H, Sudan K, Mishra A. A smart forest-fire early detection sensory system:Another approach of utilizing wireless sensor and neural networks [C].2010 IEEE Sensors Proceedings, 2010:1900-1904.
[51]T.Celik, H.Demirel, H.Ozkaramanli. Automatic fire detection in video sequences.
[52]Thou-Ho (Chao-Ho) Chen. An Early Fire-Detection Method Based on Image Processing.2004 International Conference on Image Processing.pp.1707-1710.
[53]Tuceryan M, Jain A K. Texture Analysis Handbook Pattern Recognition and Computer Vision [M]. Singapore World Scientific 1993.235-276.
[54]Tung Truong Xuan, Kim Jong-Myon. An effective four-stage smoke-detection algorithm using video images for early fire-alarm systems [J]. Fire Safety Journal,2011,46(5):276-282.
[55]Yasar Guneri Sahin. Autonomous forest fire detection.14th Conference on fire and forest meteorology. VOL.Ⅱ,pp.2003-2012,Luso,16/20 November 1998.
[56]Yoav Freund, Robert E.Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences,1997,55(1):119-133.
[2]程娟.复杂背景下运动目标识别算法研究[D].武汉理工大学,2008.
[3]戴南.基于决策树的分类方法研究而[D].南京师范大学,2003,1：4-26.
[4]冈萨雷斯.数字图像处理(第二版)[M].北京：电子工业出版社,2007：72-74.
[5]高爱莲.基于视频的火焰检测方法研究[D].昆明理工大学,2008.
[6]郭富强,于波,汪叔华.陀螺稳定装置及其应用[M].西安：西北工业大学出版社,1995.
[7]姜超.视频序列中运动目标提取与跟踪技术的研究[D].大连理工大学,2009.
[8]刘泉志.基于视频的运动目标检测与跟踪算法研究[D].上海交通大学,2009.
[9]刘听雨.基于视频特征的火灾火焰识别算法[D].西安电子科技大学.2009.
[10]罗鹏.林火协同监测技术研究[D].中国林业科学研究院,2010.
[11]罗琴娟.森林火灾视频图像智能识别技术研究[D].北京林业大学,2009.
[12]吕普铁.基于普通CCD摄像机的火灾探测技术研究[J].哈尔滨工程大学学报,2003,2.44-48
[13]康锋.基于视觉特征的早期农林火灾检测方法的基础研究[D].浙江大学,2010：31-33.
[14]李彦斌,陈树越.早期森林火灾烟雾图像序列的分割算法[J].计算机工程与应用,2008,44(11)：203-204.
[15]练秋生,李党.融合多种特征的烟雾图像检测算法[J].光学技术,2009,35(4)：523-528.
[16]Richard O.Duda,Peter E.Hart,David G Stork.模式分类[M].北京:机械工业出版社,2003: 356-412.
[17]任海,高昌海,顾香凤.欧美森林火灾监测新技术及设备[J].林业机械与木工设备,2005,33(5)：8-11.
[18]任厚平,张永明.基于混合高斯模型定位的火灾烟雾纹理特征提取[J].微计算机信息,2005(21).
[19]饶裕平,方陆明,等.林火视频监控中烟识别方法概述[J].四川林业科技,2009,30(1)：81-85.
[20]宋广玲.一种基于CART决策树改进算法[J].哈尔滨理工大学学报,2009,2.
[21]吴雪琼,覃先林,等.我国林火监测体系现状分析[J].内蒙古林业调查设计,2010,33(3)：69-72.
[22]王涛,刘渊,等.一种基于飘动性分析的视频烟雾检测新方法[J].电子与信息学报,2011,33(5)：1024-1029.
[23]西奥多里蒂斯等.模式识别(中文第二版)[M].北京：电子工业出版社,2004：106-131.
[24]肖刚.国内外森林防火技术现状及趋势探讨[D].天津大学,2005,(11)：3-65.
[25]徐鹏,任波.基于单高斯模型的森林火灾烟图像目标检测[J].计算机与现代化,2009,(2)：18-20.
[26]闫厚,倪金生等.基于网络摄像机的森林火灾烟火自动识别技术探讨[J].森林防火,2007(2)：25-28.
[27]姚志芳.加强森林防火教育的重要性及措施[J].现代农业科技,2008(14)：351-353.
[28]袁非牛.基于运动累积和半透明的视频烟雾探测模型[J].数据采集与处,2007,22(4)：396-399.
[29]郑焕能.森林防火[M].哈尔滨：东北林业大学出版社,1994.
[30]钟平.机载电子稳像技术研究[D].中国科学院长春光机所.2003.
[31]朱娟娟.电子稳像理论及其应用研究[D].西安电子科技大学,2009：3-7.
[32]Bigley W J.Optimal motion stabilization control of an electro-optiacl sight system[C].Proc.SPIE, 1989.Ⅳ:116-120.
[33]B.K.P.Hom and B.Schunk. Determining Optical Flow, Artificial Intelligence, Vol.17,1981: 185-204.
[34]B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin, "Flame detection in video using hidden markov models," ICIP'05, pp.1230-1233 [2005].
[35]B. U. Toreyin, R. G. Cinbis. Fire detection in infrared video using wavelet analysis, Optical Engineering 46(6).
[36]B.Vagvolgyi,I.Kopilovic and T.Sziranyi, Application of panoramic annular lens for motion analysis tasks:surveillance and smoke detection[C], Proceedings of 15th International Conference on Pattern Recognition,2000, Volume 4,3-7 Sept.2000:714-717.
[37]Damir Krstinic, Darko Stipanicev, Toni Jakovcevic. Histogram-based smoke segmentation in forest fire detection system[J]. Information Technology And Control,2009,38(3):237-244.
[38]F. Vella, A. Castorina, M. Mancuso, G. Messina. Digital image stabilization by adaptive block motion vectors filtering[J]. IEEE Transactions on ConsumerElectronics,2002,48(3):796-801.
[39]Gomez-Rodriguez F, Arrue B C, Ollero A. Smoke monitoring and measurement using image processing[C]. Application to forest fires. In. Proc. of SPIE, Automatic Target Recognition XIII, 2003,5094:404-411.
[40]H. J. Grech-Cini. Smoke Detection[J]. US Patent No. US6844818B2.
[41]Ham SunJae, Ko Byoung-Chul, Nam Jae-Yea. Vision based forest smoke detection using analyzing of temporal patterns of smoke and their probability models [C]. Proceedings of SPIE-The International Society for Optical Engineering,2011,78-77.
[42]Huang J C. Automatic feature-based global motion estimation in video sequences. IEEE,2004, 50(3):911-915.
[43]K.Sauer and B.Schwartz. Efficient block motion estimation using integral projections. IEEE Trans. on Circuits and Systems for Video Technology,1996,6(5):513-518
[44]Morimoto C, Chellappa R. Fast electronic digital image stabilization for off-road navigation. IEEE Journal of Real-time Imaging,1996(2):285-296.
[45]Myrheim J, Rue H. New algorithms for maximum entropy image restoration[J]. CVFIP,1992,54(3): 223-238.
[46]Ohta Y, Kandae T, Sakai T. Color information for region segmentation. Computer Graphics Image Process,13:pp.222-241,1980.
[47]Paulo Vinicius Koerich Borges. Efficient visual fire detection applied for video retrieval.
[48]R M Harliek, D Shnagmugna. Textural Feature for Image Classification, IEEE Transaction on Systems, Mna, Cybenreties, Vol.3(6), PP.610-621,1973.
[49]Surit S, Chatwiriya W. Forest fire smoke detection in video based on digital image processing approach with static and dynamic characteristic analysis [C].2011 First ACIS/JNU International Conference on CNSI Proceedings,2011:35-39.
[50]Soliman H, Sudan K, Mishra A. A smart forest-fire early detection sensory system:Another approach of utilizing wireless sensor and neural networks [C].2010 IEEE Sensors Proceedings, 2010:1900-1904.
[51]T.Celik, H.Demirel, H.Ozkaramanli. Automatic fire detection in video sequences.
[52]Thou-Ho (Chao-Ho) Chen. An Early Fire-Detection Method Based on Image Processing.2004 International Conference on Image Processing.pp.1707-1710.
[53]Tuceryan M, Jain A K. Texture Analysis Handbook Pattern Recognition and Computer Vision [M]. Singapore World Scientific 1993.235-276.
[54]Tung Truong Xuan, Kim Jong-Myon. An effective four-stage smoke-detection algorithm using video images for early fire-alarm systems [J]. Fire Safety Journal,2011,46(5):276-282.
[55]Yasar Guneri Sahin. Autonomous forest fire detection.14th Conference on fire and forest meteorology. VOL.Ⅱ,pp.2003-2012,Luso,16/20 November 1998.
[56]Yoav Freund, Robert E.Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences,1997,55(1):119-133.