一种分步的融合时空信息的背景建模

储珺; 杨樊; 张桂梅; 汪凌峰

doi:10.3724/SP.J.1004.2014.00731

一种分步的融合时空信息的背景建模

doi: 10.3724/SP.J.1004.2014.00731

1.
南昌航空大学计算机视觉研究所南昌 330063;
2.
中国科学院自动化研究所北京 100190

基金项目:

国家重点基础研究发展计划（973计划）（2009CB320902），国家自然科学基金（61263046），中国航天科技集团公司航天科技创新基金（CASC201102）资助

详细信息

作者简介:
杨樊南昌航空大学计算机视觉研究所硕士研究生.主要研究方向为目标检测与智能视频监控.E-mail：693339173@qq.com

计量
- 文章访问数: 1296
- HTML全文浏览量: 74
- PDF下载量: 707
- 被引次数: 0
出版历程
- 收稿日期: 2012-08-30
- 修回日期: 2012-12-24
- 刊出日期: 2014-04-20

A Stepwise Background Subtraction by Fusion Spatio-temporal Information

1.
Institute of Computer Vision, Nanchang Hangkong University, Nanchang 330063;
2.
Institute of Automation, Chinese Academy of Sciences

Funds:

Supported by National Basic Research Program of China (973 Program) (2009CB320902), National Natural Science Foundation of China (61263046), and the Aerospace Science and Technology Innovation Fund of China (CASC201102)

摘要

摘要: 自然场景中的光照突变和树枝、水面等不规则运动是背景建模的主要困难. 针对该问题，提出一种分步的融合时域信息和空域信息的背景建模方法. 在时域，采用具有光照不变性的颜色空间表征时域信息，并提出对噪声和光照突变具有较好适应性的码字聚类准则和自适应背景更新策略，构造了对噪声和光照突变具有较好适应性的时域信息背景模型. 在空域，通过采样将测试序列图像分成两幅子图，而后利用时域模型检测其中一幅子图，并将检测结果作为另一幅子图的先验信息，同时采用马尔科夫随机场（Markov random field，MRF）对其加以约束，最终检测其状态. 在多个测试视频序列上的实验结果表明，本文背景模型对于自然场景中的光照突变和不规则运动具有较好的适应性.
- 时空背景模型 /
- 前景检测 /
- 马尔科夫随机场 /
- 码本
Abstract: In a natural scene, it is difficult to create a background model for the presence of illumination variation and irregular motions including waving trees, rippling water, etc. This paper proposes a new stepwise algorithm by fusing spatio-temporal information. In the time domain, we characterize the temporal information in the color space which is invariant to photometric changing. On this basis, we propose a clustering criterion of codeword which is adaptive to noise and illumination variation, and present a novel adaptive background updating strategy. Then a temporal information background model which has a better adaptability to noise and photometric invariants is constructed. In the spatial domain, we first divide the test frame into two sub-images by sampling and then utilize temporal information to detect one of them. Furthermore, we regard the detection results as priori information of the other sub-image and adopt Markov random field to restrict it simultaneously, then detect its state. Extensive experiments are conducted on several test video sequences. Compared with the mixture of Gaussians (MOG), standard codebook model (SCBM), and improved codebook model (ICBM), the results show that out algorithm has better adaptability to the illumination variation and irregular movement in natural scenes.
- Spatio-temporal background model /
- foreground detection /
- Markov random field (MRF) /
- codebook

HTML全文

参考文献(21)

[1]	Qi Yu-Juan, Wang Yan-Jiang, Li Yong-Ping. Memory-based Gaussian mixture background modeling. Acta Automatica Sinica, 2010, 36(11): 1520-1526(齐玉娟, 王延江, 李永平. 基于记忆的混合高斯背景建模. 自动化学报, 2010, 36(11): 1520-1526)
[2]	Wang Yong-Zhong, Liang Yan, Pan Quan, Cheng Yong-Mei, Zhao Chun-Hui. Spatiotemporal background modeling based on adaptive mixture of Gaussians. Acta Automatica Sinica, 2009, 35(4): 371-378(王永忠, 梁彦, 潘泉, 程咏梅, 赵春晖. 基于自适应混合高斯模型的时空背景建模. 自动化学报, 2009, 35(4): 371-378)
[3]	Wren C R, Azarbayejani A, Darrell T, Pentland A P. Pfinder: real-time tracking of the human body. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(7): 780-785
[4]	Horprasert T, Harwood D, Davis L S. A statistical approach for real-time robust background subtraction and shadow detection. In: Proceedings of the 12th International Conference on Computer Vision. Kerkyra, Greece: IEEE, 1999. 1-19
[5]	Stauffer C, Grimson W E L. Adaptive background mixture models for real-time tracking. In: Proceedings of the 14th International Conference on Computer Vision and Pattern Recognition. Fort Collins, USA: IEEE, 1999. 246-252
[6]	Chen G, Yu Z Z, Wen Q, Yu Y Q. Improved Gaussian mixture model for moving object detection. Artificial Intelligence and Computational Intelligence, 2011, 7002: 179-186
[7]	Kim K, Chalidabhongse T H, Harwood D, Davis L. Real-time foreground-background segmentation using codebook model. Real-Time Imaging, 2005, 11(3): 172-185
[8]	Zhao Xu-Dong, Liu Peng, Tang Jiang-Long, Liu Jia-Feng. Background modeling adaptive to outdoor illumination variation and foreground detection approach. Acta Automatica Sinica, 2011, 37(8): 915-922(赵旭东, 刘鹏, 唐降龙, 刘家锋. 一种适应户外光照变化的背景建模及目标检测方法. 自动化学报, 2011, 37(8): 915-922)
[9]	Yin Z Z, Collins R. Belief propagation in a 3D spatio-temporal MRF for moving object detection. In: Proceedings of the 23rd Conference on Computer Vision and Pattern Recognition. Minneapolis, MN, USA: IEEE, 2007. 1-8
[10]	Migdal J, Grimson W E L. Background subtraction using Markov thresholds. In: Proceedings of the 7th Workshop on Application of Computer Vision. Breckenridge, USA: IEEE, 2005. 58-65
[11]	Wu M J, Peng X R. Spatio-temporal context for codebook-based dynamic background subtraction. AEU-International Journal of Electronics and Communication, 2010, 64(8): 739-747
[12]	Burghouts G J, Geusebroek J M. Performance evaluation of local colour invariants. Computer Vision and Image Understanding, 2009, 113(1): 48-62
[13]	Li S Z. Markov random field modeling in computer vision. In: Proceedings of the 3rd European Conference on Computer Vision. Secaucus, NJ: Springer-Verlag, 1995. 361-370
[14]	Besag J. Spatial interaction and the statistical analysis of lattice systems. Journal of the Royal Statistical Society, Series B (Methodological), 1974, 36(2): 192-236
[15]	Szeliski R, Zabih R, Scharstein D, Veksler O, Kolmogorov V, Agarwala A, Tappen M, Rother C. A comparative study of energy minimization methods for Markov random fields with smoothness-based priors. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(6): 1068-1080
[16]	Besag B J. On the statistical analysis of dirty picture. Journal of the Royal Statistical Society, Series B (Methodological), 1986, 48(3): 259-302
[17]	Felzenszwalb P F, Huttenlocher D P. Efficient belief propagation for early vision. International Journal of Computer Vision, 2006, 70(1): 41-54
[18]	Boykov Y, Veksler O, Zabih R. Fast approximate energy minimization via graph cuts. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(11): 1222-1239
[19]	Tao Lin-Mi, Wang Qi-Fan, Di Hui-Jun. Markov random field in visual information processing. Journal of Image and Graphics, 2009, 14(9): 1705-1711(陶霖密, 王奇凡, 邸慧军. 视觉信息处理中的马尔可夫随机场. 中国图象图形学报, 2009, 14(9): 1705-1711)
[20]	Heikkila M, Pietikainen M. A texture-based method for modeling the background and detecting moving objects. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(4): 657-662
[21]	Chen Y T, Chen C S, Huang C R, Huang Y P. Efficient hierarchical method for background subtraction. Pattern Recognition, 2007, 40(10): 2706-2715