基于小波矩的人体行为识别系统的设计与实现
|
摘要
近年来,对于视频序列中人体运动行为的视觉分析是计算机视觉领域中日益受到重视的一个研究方向。随着时代的进步,这项技术也在不断的发展完善,在未来必将有广阔的应用前景并为社会带来巨大的经济效益。
文章首先详细介绍了当前国内外在人体行为识别领域的研究现状和发展趋势,并且分析了其在视频监控方面的应用情况。在此基础上,本文完成的研究工作及主要贡献包括以下三个方面：
在静止背景下运动目标检测和提取研究方面,总结了当前常用的几种算法的原理,提出了一种改进的基于背景减除的时间差分法,可获得较为精确的运动人体轮廓。
在运动目标跟踪研究方面,在分析了现有跟踪方法的基础上,详细介绍了本文采用的一种适合于本系统的运动目标跟踪算法,即利用卡尔曼滤波对场景中的运动对象进行实时跟踪。
在人体行为识别方面,由于人体在运动过程中动作都可以分解成以时间为线索的图像序列,本文使用了状态空间法,首先利用小波矩来对人体图像进行特征提取,然后利用隐马尔可夫模型对运动人体的行为进行识别。
通过实验发现,本文的算法具有很小的漏检率和误检率,而且具有较好的实时性。
Visual analysis of human motion from video sequences has been attached more and more importance to computer visions in the recent years. Along with the time progress, this technology also in the unceasing development and consummation, so it will certainly have the broad application prospect and bring the huge economic efficiency for the society in the future.
Firstly this paper introduces the research status and tendency of human motion recognition at home and abroad, and then analyses its application in Security Monitoring System. In this foundation, the work in the thesis makes three main contributions in the field:
On the research of the moving objects detection and extraction under a static background, we summarize several current algorithms and propose an improved algorithm. Combining the Background Subtraction and Temporal difference, then we can get more precise moving human contour.
On the research of moving objects tracking, this paper analyzes the current tracking algorithms. Then a tracking algorithm suitable for this system is introduced in detail, which uses Kalman Filter to track the moving objects real-timely in the scenes,
In the aspect of human motion recognition, because of the fact that human continuous motion can be decomposed into an image sequence based on time, state space method is applied in this paper. Firstly this paper uses wavelet moment to extract features of the human body images, and then recognizes the moving human bodies activity based on Hidden Markov Model.
Through the experiment, the algorithm has very small leak-examining and mistake-examining-rate, moreover it has good real time.
文章首先详细介绍了当前国内外在人体行为识别领域的研究现状和发展趋势,并且分析了其在视频监控方面的应用情况。在此基础上,本文完成的研究工作及主要贡献包括以下三个方面：
在静止背景下运动目标检测和提取研究方面,总结了当前常用的几种算法的原理,提出了一种改进的基于背景减除的时间差分法,可获得较为精确的运动人体轮廓。
在运动目标跟踪研究方面,在分析了现有跟踪方法的基础上,详细介绍了本文采用的一种适合于本系统的运动目标跟踪算法,即利用卡尔曼滤波对场景中的运动对象进行实时跟踪。
在人体行为识别方面,由于人体在运动过程中动作都可以分解成以时间为线索的图像序列,本文使用了状态空间法,首先利用小波矩来对人体图像进行特征提取,然后利用隐马尔可夫模型对运动人体的行为进行识别。
通过实验发现,本文的算法具有很小的漏检率和误检率,而且具有较好的实时性。
Visual analysis of human motion from video sequences has been attached more and more importance to computer visions in the recent years. Along with the time progress, this technology also in the unceasing development and consummation, so it will certainly have the broad application prospect and bring the huge economic efficiency for the society in the future.
Firstly this paper introduces the research status and tendency of human motion recognition at home and abroad, and then analyses its application in Security Monitoring System. In this foundation, the work in the thesis makes three main contributions in the field:
On the research of the moving objects detection and extraction under a static background, we summarize several current algorithms and propose an improved algorithm. Combining the Background Subtraction and Temporal difference, then we can get more precise moving human contour.
On the research of moving objects tracking, this paper analyzes the current tracking algorithms. Then a tracking algorithm suitable for this system is introduced in detail, which uses Kalman Filter to track the moving objects real-timely in the scenes,
In the aspect of human motion recognition, because of the fact that human continuous motion can be decomposed into an image sequence based on time, state space method is applied in this paper. Firstly this paper uses wavelet moment to extract features of the human body images, and then recognizes the moving human bodies activity based on Hidden Markov Model.
Through the experiment, the algorithm has very small leak-examining and mistake-examining-rate, moreover it has good real time.
引文
[1]G.Johansson. Visual motion perception Scientific American.76-88 November1976
[2]Guo Y, Xu G and Tsuji S. Understanding human motion patterns. In:Proc International Conference on Pattern Recognition, Jerusalem, Israel,1994,325-329.
[3]赵宏明.水利行业视频监控系统研究与应用[硕士学位论文].太原理工大学.20050501
[4]Gonzalez R, Woods R. Digital image managing [M]. electron industry publishing company,2002
[5]H.Sidenbladh, Probabilistic Tracking and Reconstruction of 3D Human Motion in Monocular Video Sequences, PHD Thesis, Department of Numerical Analysis and Computer Science, KTH, Sweden,2001.
[6]王亮,胡卫明,谭铁牛.人运动的视觉分析综述.计算机学报,2002,25
[7]McKenna S et al, Tracking groups of people. Computer Vision and Image Understanding,2000,80 (1):42-56.
[8]Lipton A, Fujiyoshi H and Patil R. Moving target classification and tracking from real-time video. In:Proc IEEE Workshop on Applications of Computer Vision, Princeton, NJ,1998,8-14.
[9]章毓晋.图像工程(下册)：图像理解与计算机视觉.北京：清华大学出版社,2000.
[10]McLachlan G and Krishnan T. The EM Algorithm and Extensions. Wiley Interscience,1997.
[11]Stringa E. Morphological change detection algorithms for surveillance applications. In:British Machine Vision Conference, Bristol, UK,2000,402-411.
[12]J.K.Aggarwal, Q.Cai. Human Motion Analysis:A Review. Computer Vision and Image Understanding,1999,73(3),428-440.
[13]M.k.Leung. Y.H. Yang. First Sight:A Human Body Outline Labeling System. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(4),359-377.
[14]J.Deutscher, A.Blake, I.Reid, Articulated body motion capture by annealed particle filtering. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Hilton Head, USA,2000,volume 2,126-133.
[15]H.Sidenbladh, M.J.Black, D.J.Fleet, Stochastic tracking of 3D human figures using 2D image motion, Proceedings of the 6th European Conference on Computer Vision, Dublin, Ireland,2000,702-718.
[16]H.Sidenbladh, Probabilistic Tracking and Reconstruction of 3D Human Motion in Monocular Video Sequences, PHD Thesis, Department of Numerical Analysis and Computer Science, KTH, Sweden,2001.
[17]Wren C, Azarbayejani A, Darrell T and Pentland A. Pfinder:real-time tracking of the human body. IEEE Trans on Pattern Analysis and Machine Intelligence,1997,19 (7):780-785.
[18]Zhong Y, Jain A and Dubuisson-Jolly M. Object tracking using deformable templates. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (5): 544-549.
[19]Paragios N and Deriche R. Geodesic active contours and level sets for the detection and tracking of moving objects. IEEE Trans Pattern Analysis and Machine Intelligence, 2000,22 (3):266-280.
[20]Bertalmio M, Sapiroo G and Randll G. Morphing active contours. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (7):733-737.
[21]Peterfreund N. Robust tracking of position and velocity with Kalman snakes. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (6):564-569.
[22]Isard M and Blake A. Contour tracking by stochastic propagation of conditional density. In:Proc European Conference on Computer Vision, Cambridge,1996, 343-356.
[23]Baumberg A and Hogg D. An efficient method for contour tracking using active shape models. In:Proc IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Austin,1994,194-199.
[23]Myers C, Rabinier L and Rosenberg A. Performance tradeoffs in dynamic time warping algorithms for isolated word recognition. IEEE Trans Acoustics, Speech, and Signal Processing,1980,28 (6):623-635.
[24]Bobick A and Wilson A. A state-based technique for the summarization and recognition of gesture. In:Proc International Conference on Computer Vision, Cambridge,1995,382-388.
[25]Takahashi K, and Seki S et al. Recognition of dexterous manipulations from time varying images. In:Proc IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Austin,1994,23-28.
[26]Poritz A. Hidden Markov Models:a guided tour. In:Proc IEEE International Conference on Acoustics, Speech and Signal Processing, New York City, NY,1988, 7-13.
[27]Guo Y, Xu G and Tsuji S. Understanding human motion patterns. In:Proc International Conference on Pattern Recognition, Jerusalem, Israel,1994,325-329.
[28]Rosenblum M, Yacoob Y and Davis L. Human emotion recognition from motion using a radial basis function network architecture. In:Proc IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Austin,1994,43-49.
[29]Davis J and Bobick A. The representation and recognition of action using temporal templates. MIT Media Lab, Perceptual Computing Group, Technical report:402,1997.
[30]Rabinier L. A tutorial on Hidden Markov models and selected applications in speech recognition. In:Proc IEEE 1989,77(2):257-285.
[31]Bregler C. Learning and recognizing human dynamics in video sequences. In:Proc IEEE Conference on Computer Vision and Pattern Recognition, Puerto Rico,1997, 568-574.
[32]Campbell L and Bobick A. Recognition of human body motion using phase space constraints. In:Proc International Conference on Computer Vision, Cambridge,1995, 624-630.
[33]Haritaoglu, I. and Harwood. D; Davis,etc.,"W/sup4/:Who? When? Where? What? A real time system for detecting and tracking people," Proceedings of Third IEEE International Conference on Automatic Face and Gesture Recognition,pp.222-227,1998.
[34]Stringa,E. and Regazzoni, C.S., "Real-time video-shot detection for scene surveillance applications, "IEEE Transactions on Image Processing, vol.9, no.1, pp.69-79,Jan.2000.
[35]J.K.Aggarwal, Q.Cai. Human Motion Analysis:A Review. Computer Vision and Image Understanding,1999,73(3),428-440.
[35]Bertalmio M, Sapiroo G and Randll G. Morphing active contours. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (7):733-737.
[37]张嫣,姚耀文,唐华松,运动图像序列中多目标跟踪的研究与实现,计算机应用研究,No.1,pp.74-76,2002
[38]Hu M. Visual pattern recognition by moment invariants. Ire Trans Inf Theory,1962(8):179～187
[39]Li Y. Reforming the theory of invariant moments for pattern recognition. Pattern Recognition,1992,25(7):723-731
[40]张淼.基于矩不变量的印鉴鉴别技术的研究[硕士学位论文].西安电子科技大学.20050126
[41]张虹.基于小波变换和不变矩的图像目标识别算法研究[硕士学位论文].北京工业大学.20040402
[42]Koller, K. Danilidis, and H.Nagel, Model-based object tracking in monocular image sequence of road traffic scenes. Int. J. Computer vision, vol.10.no.3,pp.257-281, 1993.
[43]Pentland, Machine understanding of human action. In Proc.7th Int. Forum Frontier of Telecommunication Technology, Tokyo, Nov.1995.
[44]P.Huttenlocker and W. Rucklidge, Tracking nonrigid objects in complex scenes. Proc. Int. Conf. Computer Vision (ICCV), Sept.1992.
[45]Y.B. Shalom, T. Fortman, Tracking and Data Association, Academic Press,1998.
[2]Guo Y, Xu G and Tsuji S. Understanding human motion patterns. In:Proc International Conference on Pattern Recognition, Jerusalem, Israel,1994,325-329.
[3]赵宏明.水利行业视频监控系统研究与应用[硕士学位论文].太原理工大学.20050501
[4]Gonzalez R, Woods R. Digital image managing [M]. electron industry publishing company,2002
[5]H.Sidenbladh, Probabilistic Tracking and Reconstruction of 3D Human Motion in Monocular Video Sequences, PHD Thesis, Department of Numerical Analysis and Computer Science, KTH, Sweden,2001.
[6]王亮,胡卫明,谭铁牛.人运动的视觉分析综述.计算机学报,2002,25
[7]McKenna S et al, Tracking groups of people. Computer Vision and Image Understanding,2000,80 (1):42-56.
[8]Lipton A, Fujiyoshi H and Patil R. Moving target classification and tracking from real-time video. In:Proc IEEE Workshop on Applications of Computer Vision, Princeton, NJ,1998,8-14.
[9]章毓晋.图像工程(下册)：图像理解与计算机视觉.北京：清华大学出版社,2000.
[10]McLachlan G and Krishnan T. The EM Algorithm and Extensions. Wiley Interscience,1997.
[11]Stringa E. Morphological change detection algorithms for surveillance applications. In:British Machine Vision Conference, Bristol, UK,2000,402-411.
[12]J.K.Aggarwal, Q.Cai. Human Motion Analysis:A Review. Computer Vision and Image Understanding,1999,73(3),428-440.
[13]M.k.Leung. Y.H. Yang. First Sight:A Human Body Outline Labeling System. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(4),359-377.
[14]J.Deutscher, A.Blake, I.Reid, Articulated body motion capture by annealed particle filtering. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Hilton Head, USA,2000,volume 2,126-133.
[15]H.Sidenbladh, M.J.Black, D.J.Fleet, Stochastic tracking of 3D human figures using 2D image motion, Proceedings of the 6th European Conference on Computer Vision, Dublin, Ireland,2000,702-718.
[16]H.Sidenbladh, Probabilistic Tracking and Reconstruction of 3D Human Motion in Monocular Video Sequences, PHD Thesis, Department of Numerical Analysis and Computer Science, KTH, Sweden,2001.
[17]Wren C, Azarbayejani A, Darrell T and Pentland A. Pfinder:real-time tracking of the human body. IEEE Trans on Pattern Analysis and Machine Intelligence,1997,19 (7):780-785.
[18]Zhong Y, Jain A and Dubuisson-Jolly M. Object tracking using deformable templates. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (5): 544-549.
[19]Paragios N and Deriche R. Geodesic active contours and level sets for the detection and tracking of moving objects. IEEE Trans Pattern Analysis and Machine Intelligence, 2000,22 (3):266-280.
[20]Bertalmio M, Sapiroo G and Randll G. Morphing active contours. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (7):733-737.
[21]Peterfreund N. Robust tracking of position and velocity with Kalman snakes. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (6):564-569.
[22]Isard M and Blake A. Contour tracking by stochastic propagation of conditional density. In:Proc European Conference on Computer Vision, Cambridge,1996, 343-356.
[23]Baumberg A and Hogg D. An efficient method for contour tracking using active shape models. In:Proc IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Austin,1994,194-199.
[23]Myers C, Rabinier L and Rosenberg A. Performance tradeoffs in dynamic time warping algorithms for isolated word recognition. IEEE Trans Acoustics, Speech, and Signal Processing,1980,28 (6):623-635.
[24]Bobick A and Wilson A. A state-based technique for the summarization and recognition of gesture. In:Proc International Conference on Computer Vision, Cambridge,1995,382-388.
[25]Takahashi K, and Seki S et al. Recognition of dexterous manipulations from time varying images. In:Proc IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Austin,1994,23-28.
[26]Poritz A. Hidden Markov Models:a guided tour. In:Proc IEEE International Conference on Acoustics, Speech and Signal Processing, New York City, NY,1988, 7-13.
[27]Guo Y, Xu G and Tsuji S. Understanding human motion patterns. In:Proc International Conference on Pattern Recognition, Jerusalem, Israel,1994,325-329.
[28]Rosenblum M, Yacoob Y and Davis L. Human emotion recognition from motion using a radial basis function network architecture. In:Proc IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Austin,1994,43-49.
[29]Davis J and Bobick A. The representation and recognition of action using temporal templates. MIT Media Lab, Perceptual Computing Group, Technical report:402,1997.
[30]Rabinier L. A tutorial on Hidden Markov models and selected applications in speech recognition. In:Proc IEEE 1989,77(2):257-285.
[31]Bregler C. Learning and recognizing human dynamics in video sequences. In:Proc IEEE Conference on Computer Vision and Pattern Recognition, Puerto Rico,1997, 568-574.
[32]Campbell L and Bobick A. Recognition of human body motion using phase space constraints. In:Proc International Conference on Computer Vision, Cambridge,1995, 624-630.
[33]Haritaoglu, I. and Harwood. D; Davis,etc.,"W/sup4/:Who? When? Where? What? A real time system for detecting and tracking people," Proceedings of Third IEEE International Conference on Automatic Face and Gesture Recognition,pp.222-227,1998.
[34]Stringa,E. and Regazzoni, C.S., "Real-time video-shot detection for scene surveillance applications, "IEEE Transactions on Image Processing, vol.9, no.1, pp.69-79,Jan.2000.
[35]J.K.Aggarwal, Q.Cai. Human Motion Analysis:A Review. Computer Vision and Image Understanding,1999,73(3),428-440.
[35]Bertalmio M, Sapiroo G and Randll G. Morphing active contours. IEEE Trans Pattern Analysis and Machine Intelligence,2000,22 (7):733-737.
[37]张嫣,姚耀文,唐华松,运动图像序列中多目标跟踪的研究与实现,计算机应用研究,No.1,pp.74-76,2002
[38]Hu M. Visual pattern recognition by moment invariants. Ire Trans Inf Theory,1962(8):179～187
[39]Li Y. Reforming the theory of invariant moments for pattern recognition. Pattern Recognition,1992,25(7):723-731
[40]张淼.基于矩不变量的印鉴鉴别技术的研究[硕士学位论文].西安电子科技大学.20050126
[41]张虹.基于小波变换和不变矩的图像目标识别算法研究[硕士学位论文].北京工业大学.20040402
[42]Koller, K. Danilidis, and H.Nagel, Model-based object tracking in monocular image sequence of road traffic scenes. Int. J. Computer vision, vol.10.no.3,pp.257-281, 1993.
[43]Pentland, Machine understanding of human action. In Proc.7th Int. Forum Frontier of Telecommunication Technology, Tokyo, Nov.1995.
[44]P.Huttenlocker and W. Rucklidge, Tracking nonrigid objects in complex scenes. Proc. Int. Conf. Computer Vision (ICCV), Sept.1992.
[45]Y.B. Shalom, T. Fortman, Tracking and Data Association, Academic Press,1998.