基于光流块统计特征的视频异常行为检测算法

上海交通大学学报（自然版） ›› 2015, Vol. 49 ›› Issue (08): 1199-1204.

• 自动化技术、计算机技术 • 上一篇下一篇

基于光流块统计特征的视频异常行为检测算法

余昊a，孙锬锋a,b，蒋兴浩a,b

(上海交通大学 a. 电子信息与电气工程学院； b. 信息内容分析技术国家工程实验室，上海 200240)

收稿日期:2014-06-18 出版日期:2015-08-31 发布日期:2015-08-31
基金资助:
国家自然科学基金项目（61272249，61272439），上海市科委国际研究合作项目（12510708500），国家教委博士点专项基金项目（20120073110053），软件工程国家实验室开放研究基金项目（SKLSE20120912）资助

Video Anomaly Detection Based on Statistic Feature of Optical Flow Block

YU Haoa，SUN Tanfenga,b，JIANG Xinghaoa,b

(a. School of Electronic Information and Electrical Engineering； b. National Engineering Laboratory on Information Content Analysis Techniques, Shanghai Jiaotong University, Shanghai 200240, China)

Received:2014-06-18 Online:2015-08-31 Published:2015-08-31

摘要/Abstract

摘要：

摘要：提出了一种基于光流块统计特征的视频异常行为检测算法.该算法首先对训练集视频序列的光流场进行分块及预处理，而后提取光流块的统计特征，所提取的块统计特征同时包括了光流块的幅度信息和相位信息，通过训练集得到的光流块统计特征训练出对应的正常行为的高斯混合模型（GMM）.测试集通过同样的方式提取光流块统计特征，通过计算所提取统计特征以多大的概率属于GMM判定所检测光流块的异常程度.实验结果表明，该算法能够在一定程度上解决运动物体一致性和部分遮挡问题，并提高了异常行为检测的准确率.

关键词: 异常行为检测, 光流块, 统计特征, 预处理, 高斯混合模型

Abstract:

Abstract： An anomaly detection algorithm based on the statistic feature of optical flow block was proposed. First, the whole optical flow field of training video sequences were obtained. Then, each optical flow field was divided into blocks and each block was preprocessed in order to extract the statistic feature considering both magnitude and phase information of the block. The Gaussian mixture model (GMM) was employed to establish the probability model of normal behaviors by feeding the statistic feature into it. The abnormal degree of the optical flow block was judged by the output posterior probability of the GMM probabilistic model. The experimental results show that the method proposed considers both the consistency information of moving objects and the partial occlusion issue, at the same time, improves the accuracy of anomaly detection.

Key words: anomaly detection, optical flow block, statistic feature, preprocessing, Gaussian mixture model

中图分类号:

TP 391

余昊a，孙锬锋a,b，蒋兴浩a,b. 基于光流块统计特征的视频异常行为检测算法[J]. 上海交通大学学报（自然版）, 2015, 49(08): 1199-1204.

YU Haoa，SUN Tanfenga,b，JIANG Xinghaoa,b. Video Anomaly Detection Based on Statistic Feature of Optical Flow Block[J]. Journal of Shanghai Jiaotong University, 2015, 49(08): 1199-1204.

参考文献

［1］Stauffer C， Grimson W E L.Learning patterns of activity using realtime tracking［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence，2000，22(8)：747757.

［2］Morris B T，Trivedi M M.Learning, modeling and classification of vehicle track patterns from live video［J］. IEEE Transactions on Intelligent Transportation Systems，2008，9(3)：425437.

［3］Basharat A, Gritai A, Shah M.Learning object motion patterns for anomaly detection and improved object detection［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Anchorage：IEEE， 2008： 18.

［4］Jiang F, Yuan J, Tsaftaris S A, et al. Video anomaly detection in spatiotemporal context［C］∥Proceedings of the International Conference on Image Processing. Hong Kong： IEEE， 2010： 705708.

［5］Zhang T, Lu H, Li S Z. Learning semantic scene models by object classification and trajectory clustering［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Miami：IEEE，2009： 19401947.

［6］Mehran R, Oyama A, Shah M.Abnormal crowd behavior detection using social force model［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Miami： IEEE， 2009： 935942.

［7］Mahadevan V, Li W, Bhalodia V, et al. Anomaly detection in crowded scenes［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. San Francisco： IEEE， 2010： 19751981.

［8］Xu D, Wu X, Song D, et al. Hierarchical activity discovery within spatiotemporal context for video anomaly detection［C］∥Image Processing (ICIP), 2013 20th IEEE International Conference on. Melbourne： IEEE, 2013： 35973601.

［9］Liu C. Beyond pixels: Exploring new representations and applications for motion analysis［D］. Cambridge：Massachusetts Institute of Technology，2009.

［10］Reynolds D A, Quatieri T F, Dunn R B. Speaker verification using adapted Gaussian mixture models［J］. Digital signal processing，2000，10(1)：1941.

［11］Zivkovic Z.Improved adaptive Gaussian mixture model for background subtraction［C］∥Proceedings of the International Conference on Pattern Recognition. Cambridge： IEEE， 2004： 2831.

［12］Kim J， Grauman K.Observe locally, infer globally: A spacetime MRF for detecting abnormal activities with incremental updates［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Miami： IEEE， 2009： 29212928.

[1]	贾晓立，江晓波，蒋三新，刘佩林. 利用结构特征的语音压缩感知重建算法[J]. 上海交通大学学报（自然版）, 2017, 51(9): 1111-1116.
[2]	成谢锋,傅女婷. 心音身份识别综述[J]. 上海交通大学学报（自然版）, 2014, 48(12): 1745-1750.
[3]	陈明. 桥梁预警系统的数据预处理 [J]. 上海交通大学学报（自然版）, 2012, 46(10): 1680-1686.

基于光流块统计特征的视频异常行为检测算法

Video Anomaly Detection Based on Statistic Feature of Optical Flow Block

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价