基于光流模值变化频率的群体骚乱行为检测方法

doi:10.11896/j.issn.1002-137X.2016.05.054

摘要/Abstract

摘要： 群体骚乱行为突发性强、破坏性大,是视频监控关注的重点。研究群体骚乱行为的智能检测方法有助于提高视频监控的智能化水平。现有行为检测方法在检测群体骚乱行为时虚警率较高,实用性较差。因此,依据群体骚乱行为发生时光流模值变化大的特性,提出了一种基于光流模值变化频率的群体骚乱行为检测方法。该方法首先计算视频中每一帧图像上各像素点的光流；然后自适应求取反映像素点光流变化大小的二值映射图；接着分区块计算视频片段上的光流变化频率直方图,构建行为描述子；最后采用线性支持向量机进行特征训练与分类。实验表明,所提方法在检测群体骚乱行为时虚警率和漏警率低、识别率高,可广泛用于智能视频监控领域。

关键词: 群体骚乱行为,行为检测,光流,支持向量机

Abstract: Group riot activity is the focus of video surveillance,which is emergent and destructive.To detect the group riot activity intelligently is helpful for improving the intelligence level of video surveillance.The common activity detection methods have high false alarm rate while detecting group riot activity from videos.In this paper,a detection method for group riot activity was proposed based on change frequency of optical flow’s magnitude.Firstly, the optical flow of each pixel on each frame in a video is calculated.Secondly,a binary map is obtained to reflect the changes of optical flow aptively.Then an activity descriptor by dividing an image into several blobs and the change frequency histogram of optical flow is calculated independently.Finally,the activity features are trained and classified by using a linear support vector machine.Experiments show that the new method has low false alarm rate,low missed alarm rate and high genuine acceptance rate,while detecting group riot activity.So it can be widely used in the field of intelligent video surveillance.

Key words: Group riot activity,Activity detection,Optical flow,SVM

林杰,林拉. 基于光流模值变化频率的群体骚乱行为检测方法[J]. 计算机科学, 2016, 43(5): 283-287. https://doi.org/10.11896/j.issn.1002-137X.2016.05.054

LIN Jie and LIN La. Detection Method for Group Riot Activity Based on Change Frequency of Optical Flow’s Magnitude[J]. Computer Science, 2016, 43(5): 283-287. https://doi.org/10.11896/j.issn.1002-137X.2016.05.054

参考文献

[1] Austin O T A.Human activity analysis:a review[J].ACM Computing Surveys,2011,43(3):194-218
[2] Thonnat M.Group behavior recognition with multiple cameras[C]∥IEEE Winter Conference on Applications of Computer Vision.2002:1-8
[3] Hu Zhi-lan,Jiang fan,Wang Gui-jin,et al.Anomaly detection based on motion direction[J].Acta Automatica Sinica,2008,4 (11):1348-1357(in Chinese) 胡芝兰,江帆,王贵锦,等.基于运动方向的异常行为检测[J].自动化学报,2008,34(11):1348-1357
[4] Liu J,Kuipers B,Savarese S.Recognizing human actions by attributes[J].University of Michigan,2011,42(7):3337-3344
[5] Han Lei,Li Jun-feng,JiaYun-de.Human Interaction Recognition Using Spatio-Temporal Words[J].Chinese Journal of Computers,2010,3(4):776-784(in Chinese) 韩磊,李君峰,贾云得.基于时空单词的两人交互行为识别方法[J].计算机学报,2010,33(4):776-784
[6] Niebles J C,Feifei L.A hierarchical model of shape and appea-rance for human action classification[C]∥IEEE Conference on Computer Vision and Pattern Recognition.2007:1-8
[7] Yuan He-jin,Wang Cui-ru.Human action recognition using Markov random walk based scmi-supervised learning[J].Journal of Computer-Aided Design & Compueter Graphics,2011,3(10):1749-1757(in Chinese) 袁和金,王翠茹.人体行为识别的Markov随机游走半监督学习方法[J].计算机辅助设计与图形学学报,2011,23(10):1749-1757
[8] Wang Chuan-xu,Liu Yun,Li Wan-qing.Research of Unsupervised Posture Modeling and Action Recognition Based on Spatial-Temporal Interesting Points [J].Acta Electronica Sinica,2011,9(8):1751-1756(in Chinese) 王传旭,刘云,厉万庆.基于时空特征点的非监督姿态建模和行为识别的算法研究[J].电子学报,2011,39(8):1751-1756
[9] Mohamed B,Kaaniche,Franios.Recognizing Gestures by Learning Local Motion Signatures of HOG Descriptors[J].Arn Analy and Mahn Nllgn Ranaon on,2012,34(11):2247-2258
[10] Dollar P,Rabaud V,Cottrell G,et al.Behavior recognition via sparse spatio-temporal features[C]∥2005 IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance.2005:65-72
[11] Du You-tian,Chen Feng,Xu Wen-li.Approach to Human Activity Multi-scale Analysis and Recognition Based on Multi-layer DynamicBayesian Network[J].Acta Automatica Sinica,2009,5(3):225-232(in Chinese) 杜友田,陈峰,徐文立.基于多层动态贝叶斯网络的人的行为多尺度分析及识别方法[J].自动化学报,2009,35(3):225-232
[12] Saad A,Mubarak S.Human action recognition in videos using kinematic features and multiple instance learning[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(2):288-303
[13] Rodriguez M,Ali S,Kanade T.Tracking in unstructured crowdedscenes[C]∥2009 IEEE 12th International Conference on Computer Vision.IEEE,2009:1389-1396
[14] Basri R,Hassner T,Zelnik-Manor L.Approximate nearest subspace search[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(2):266-278
[15] Ke Y,Sukthankar R,Hebert M.Event Detection in CrowdedVideos[C]∥IEEE 11th International Conference on Computer Vision,2007(ICCV 2007).IEEE,2007:1-8
[16] Kellokumpu V,Zhao G,Pietikinen M.Human Activity Recognition Using a Dynamic Texture Based Method[J].In BMVC,2008
[17] Bermejo N E,Deniz S O,Bueno G G,et al.Violence detection in video using computer vision techniques[C]∥International Conference on Computer Analysis of Images and Patterns.2011:332-339
[18] Orit K G,Tal H,Lior W,et al.The Action Similarity Labeling Challenge[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(3):615-621
[19] Yeffet L,Wolf L.Local Trinary Patterns for Human ActionRecognition[C]∥IEEE International Conference on Computer Vision,2009:492-497
[20] Hayashi K,Seki M,Hirai T,et al.Real-time violent action detector for elevator[J].Optomechatronic Machine Vision,2005:60510R-60510R-8
[21] Horn B K P,Schunck B G.Determining Optical Flow[J].Artificial Intelligence,1980,17(81):185-203
[22] Kim H,Pang S,Je H,et al.Constructing support vector machine ensemble[J].Pattern Recognition,2003,36(12):2757-2767
[23] Wang Yuan-yuan,Wang Bin.The Conditional Random Fields Me-thods for Human Action Recognition[J].Journal of Chongqing University of Technology(Natural Science),2013,7(6):93-99,105(in Chinese) 王媛媛,王斌.人体行为识别的条件随机场方法[J].重庆理工大学学报(自然科学),2013,27(6):93-99,105

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed