计算机科学 ›› 2021, Vol. 48 ›› Issue (7): 199-205.doi: 10.11896/jsjkx.200800146

• 计算机图形学&多媒体 • 上一篇    下一篇

基于自反馈最优子类挖掘的视频异常检测算法

侯春萍, 赵春月, 王致芃   

  1. 天津大学电气与信息工程学院 天津300072
  • 收稿日期:2020-08-22 修回日期:2020-10-24 出版日期:2021-07-15 发布日期:2021-07-02
  • 通讯作者: 王致芃(zpwang@tju.edu.cn)
  • 基金资助:
    国际合作与交流NFSC项目(61520106002);国家自然科学基金(61731003)

Video Abnormal Event Detection Algorithm Based on Self-feedback Optimal Subclass Mining

HOU Chun-ping, ZHAO Chun-yue, WANG Zhi-peng   

  1. School of Electrical and Information Engineering,Tianjin University,Tianjin 300072,China
  • Received:2020-08-22 Revised:2020-10-24 Online:2021-07-15 Published:2021-07-02
  • About author:HOU Chun-ping,born in 1957,Ph.D,professor,Ph.D supervisor.Her main research interests include wireless communications and 3D imaging.(hcp@tju.edu.cn)
    WANG Zhi-peng,born in 1992,Ph.D.His main research intersets include video and image anomly detection.
  • Supported by:
    projects of International Cooperation and Exchanges NFSC(61520106002) and National Natural Science Foundation of China(61731003).

摘要: 视频异常检测算法是视频处理领域的研究热点之一,用于检测视频中是否包含异常事件。然而,由于没有异常样本参与训练过程,且异常样本与正常样本之间存在一定程度的相似性,因此很难设计出一种有辨识力的异常检测模型。为了解决上述问题,文中首先提出了一种基于相似度保持和样本恢复的特征选择方法,该方法能够保留正常样本的相似关系,进而可以学习到能够准确描述正常事件的特征。其次,将异常事件检测任务转化为分类任务,并提出了一种自反馈最优子类挖掘方法来获得最优分类器。如果一个测试样本被所有分类器判断为异常,则该样本最终将被判定为异常。在公共视频数据集(Avenue数据集、UCSD Ped2数据集)上进行的大量实验的结果表明,所提异常事件检测算法可以取得很好的结果。

关键词: 视频异常事件检测, 特征选择, 一类支持向量机, 自反馈, 最优子类挖掘

Abstract: Video anomaly detection algorithm is one of the hot issues in the field of video processing,and it is used to detect whether an abnormal event is contained in the video.However,since abnormal samples are not involved in the training process,and there is a certain degree of similarity between abnormal samples and normal samples,it is difficult to design an abnormal detection model with discrimination.In order to solve the above problems,firstly,this paper proposes a feature selection method based on similarity preservation and sample recovery.This method can retain the similarity of normal samples,and then learn features that can accurately describe normal events.Secondly,it formalizes the abnormal event detection as classification problem,and proposes a self-feedback optimal subclass mining method to find optimal classifier.The sample will be labeled as anomaly if all classifiers label it as anomaly.Extensive experiments on public video surveillance datasets (i.e.Avenue Dataset and UCSD Ped2 Dataset) demonstrate that the proposed abnormal event detection method can achieve good results.

Key words: Feature selection, One-class SVM, Optimal subclass mining, Self-feedback, Video abnormal event detection

中图分类号: 

  • TN911.73
[1]SHI Y,YI Y,ZHANG Q,et al.Kernel null-space-based abnormal event detection using hybrid motion information[J].Journal of Electronic Imaging,2019,28(2):1-12.
[2]YU Y,SHEN W,HUANG H,et al.Abnormal event detection in crowded scenes using two sparse dictionaries with saliency[J].Journal of Electronic Imaging,2017,26(3),033013.
[3]HASAN M,CHOI J,NEUMANN J,et al.Learning temporalregularity in video sequences[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:733-742.
[4]TUDOR I R,SMEUREANU S,ALEXE B,et al.Unmasking the abnormal events in video[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2895-2903.
[5]RAVANBAKHSH M,NABI M,SANGINETO E,et al.Abnormal event detection in videos using generative adversarial nets[C]//2017 IEEE International Conference on Image Processing (ICIP).2017:1577-1581.
[6]WANG S,ZHU E,YIN J,et al.Anomaly detection in crowded scenes by sl-hof descriptor and foreground classification[C]//23rd International Conference on Pattern Recognition (ICPR).2016.
[7]WANG S,ZHU E,YIN J,et al.Video anomaly detection and localization by local motion based joint video representation and ocelm[J].Neurocomputing,2018,277(FEB):161-175.
[8]LEI Q,YE Y,L S,et al.Abnormal event detection based on multi-scale markov random field[C]//Chinese Conference on Computer Vision.2015.
[9]CUI L,LI K,CHEN J,et al.Abnormal event detection in traffic video surveillance based on local features[C]//2011 4th International Congress on Image and Signal Processing.2011:362-366.
[10]ZELNIK-MANOR L,ROSENBLUM K,ELDAR Y C.Dictionary optimization for block-sparse representations[J].IEEE Transactions on Signal Processing,2012,60(5):2386-2395.
[11]IONESCU R T,SMEUREANU S,POPESCU M,et al.Detecting abnormal events in video using narrowed normality clusters[C]//in 2019 IEEE Winter Conference on Applications of Computer Vision (WACV).2019:1951-1960.
[12]AMRAEE S,VAFAEI A,JAMSHIDI K,et al.Abnormal event detection in crowded scenes using one-class svm[J].Signal,Ima-ge and Video Processing,2018,12(6):1115-1123.
[13]LIU X,WANG L,ZHANG J,et al.Global and local structure preservation for feature selection[J].IEEE Transactions on Neural Networks and Learning Systems,2017,25(6):1083-1095.
[14]SHAO W,HE L,LU C T,et al.Online unsupervised multi-view feature selection[C]//2016 IEEE 16th International Conference on Data Mining (ICDM).2016:1203-1208.
[15]ZHANG T,JIE Y,ZHAO D,et al.Linear local tangent spacealignment and application to face recognition[J].Neurocompu-ting,2007,70(7):1547-1553.
[16]ROWEIS S T,SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290(5500),2323-2326.
[17]HE X,NIYOGI P.Locality preserving projections[C]//Ad-vances in Neural Information Processing Systems.2004:153-160.
[18]ZHOU T,ZHANG C,GONG C,et al.Multiview latent space learning with feature redundancy minimization[J].IEEE Tran-sactions on Cybernetics,2018,99:1-14.
[19]LIN Z C,LIU R,SU Z.Linearized alternating direction method with adaptive penalty for low-rank representation[J].arXiv:1109.0367,2011.
[20]LIU G,LIN Z,YAN S,et al.Robust recovery of subspace structures bylow-rank representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,35(1):171-184.
[21]TANG C,ZHU X,LIU X,et al.Cross-view local structure preserved diversity and consensus learning for multi-view unsupervised feature selection[C]//Proceedings of the AAAI Confe-rence on Artificial Intelligence.2019:5101-5108.
[22]IONESCU R T,KHAN F S,GEORGESCU M I,et al.Object-centric auto-encoders and dummy anomalies for abnormal event detection in video[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2019:7842-7851.
[23]VEDALDI A,FULKERSON B.Vlfeat:An open and portable library of computer vision algorithms[C]//Proceedings of the 18th International Conference on Multimedea 2010.Firenze,Italy,2010.
[24]DU Q,FABER V,GUNZBURGER M.Centroidal voronoi tessellations:Applications and algorithms[J].SIAM review,1999,41(4):637-676.
[25]ARTHUR D,VASSILVITSKII S.K-means++:The advantages of careful seeding[C]//Eighteenth Acm-siam Symposium on Discrete Algorithms.2007:1027-1035.
[26]LU C,SHI J,JIA J.Abnormal event detection at 150 fps in matlab[C]//Proceedings of the IEEE International Conference on Computer Vision.2013:2720-2727.
[27]DEL GIORNO A,BAGNELL J A,HEBERT M.A discriminative framework for anomaly detection in large videos[C]//European Conference on Computer Vision.Springer,2016:334-349.
[28]LUO W,LIU W,GAO S.A revisit of sparse coding based ano-maly detection in stacked rnn framework[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:341-349.
[29]MEHRAN R,OYAMA A,SHAH M.Abnormal crowd behavior detection using social force model[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009:935-942.
[30]CHONG Y S,TAY Y H.Abnormal event detection in videosusing spatiotemporal autoencoder[C]//International Sympo-sium on Neural Networks.Springer,2017:189-196.
[31]KIM J K,GRAUMAN K.Observe locally,infer globally:Aspace-time mrf for detecting abnormal activities with incremental updates[C]//IEEE Conference on Computer Vision and Pattern Recognition.2009:921-2928.
[32]MAHADEVAN V,LI W,BHALODIA V,et al.Anomaly detection in crowded scenes[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.IEEE,2010:1975-1981.
[1] 李斌, 万源.
基于相似度矩阵学习和矩阵校正的无监督多视角特征选择
Unsupervised Multi-view Feature Selection Based on Similarity Matrix Learning and Matrix Alignment
计算机科学, 2022, 49(8): 86-96. https://doi.org/10.11896/jsjkx.210700124
[2] 孙奇, 吉根林, 张杰.
基于非局部注意力生成对抗网络的视频异常事件检测方法
Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection
计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[3] 胡艳羽, 赵龙, 董祥军.
一种用于癌症分类的两阶段深度特征选择提取算法
Two-stage Deep Feature Selection Extraction Algorithm for Cancer Classification
计算机科学, 2022, 49(7): 73-78. https://doi.org/10.11896/jsjkx.210500092
[4] 康雁, 王海宁, 陶柳, 杨海潇, 杨学昆, 王飞, 李浩.
混合改进的花授粉算法与灰狼算法用于特征选择
Hybrid Improved Flower Pollination Algorithm and Gray Wolf Algorithm for Feature Selection
计算机科学, 2022, 49(6A): 125-132. https://doi.org/10.11896/jsjkx.210600135
[5] 储安琪, 丁志军.
基于灰狼优化算法的信用评估样本均衡化与特征选择同步处理
Application of Gray Wolf Optimization Algorithm on Synchronous Processing of Sample Equalization and Feature Selection in Credit Evaluation
计算机科学, 2022, 49(4): 134-139. https://doi.org/10.11896/jsjkx.210300075
[6] 孙林, 黄苗苗, 徐久成.
基于邻域粗糙集和Relief的弱标记特征选择方法
Weak Label Feature Selection Method Based on Neighborhood Rough Sets and Relief
计算机科学, 2022, 49(4): 152-160. https://doi.org/10.11896/jsjkx.210300094
[7] 李宗然, 陈秀宏, 陆赟, 邵政毅.
鲁棒联合稀疏不相关回归
Robust Joint Sparse Uncorrelated Regression
计算机科学, 2022, 49(2): 191-197. https://doi.org/10.11896/jsjkx.210300034
[8] 张叶, 李志华, 王长杰.
基于核密度估计的轻量级物联网异常流量检测方法
Kernel Density Estimation-based Lightweight IoT Anomaly Traffic Detection Method
计算机科学, 2021, 48(9): 337-344. https://doi.org/10.11896/jsjkx.200600108
[9] 杨蕾, 降爱莲, 强彦.
基于自编码器和流形正则的结构保持无监督特征选择
Structure Preserving Unsupervised Feature Selection Based on Autoencoder and Manifold Regularization
计算机科学, 2021, 48(8): 53-59. https://doi.org/10.11896/jsjkx.200700211
[10] 胡艳梅, 杨波, 多滨.
基于网络结构的正则化逻辑回归
Logistic Regression with Regularization Based on Network Structure
计算机科学, 2021, 48(7): 281-291. https://doi.org/10.11896/jsjkx.201100106
[11] 周钢, 郭福亮.
基于特征选择的高维数据集成学习方法研究
Research on Ensemble Learning Method Based on Feature Selection for High-dimensional Data
计算机科学, 2021, 48(6A): 250-254. https://doi.org/10.11896/jsjkx.200700102
[12] 丁思凡, 王锋, 魏巍.
一种基于标签相关度的Relief特征选择算法
Relief Feature Selection Algorithm Based on Label Correlation
计算机科学, 2021, 48(4): 91-96. https://doi.org/10.11896/jsjkx.200800025
[13] 滕俊元, 高猛, 郑小萌, 江云松.
噪声可容忍的软件缺陷预测特征选择方法
Noise Tolerable Feature Selection Method for Software Defect Prediction
计算机科学, 2021, 48(12): 131-139. https://doi.org/10.11896/jsjkx.201000168
[14] 张亚钏, 李浩, 宋晨明, 卜荣景, 王海宁, 康雁.
混合人工化学反应优化和狼群算法的特征选择
Hybrid Artificial Chemical Reaction Optimization with Wolf Colony Algorithm for Feature Selection
计算机科学, 2021, 48(11A): 93-101. https://doi.org/10.11896/jsjkx.210100067
[15] 董明刚, 黄宇扬, 敬超.
基于遗传实例和特征选择的K近邻训练集优化方法
K-Nearest Neighbor Classification Training Set Optimization Method Based on Genetic Instance and Feature Selection
计算机科学, 2020, 47(8): 178-184. https://doi.org/10.11896/jsjkx.190700089
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!