基于有效异常样本构造的视频异常检测算法

doi:10.13229/j.cnki.jdxbgxb20200441

摘要/Abstract

摘要：

为改善大多数异常检测算法仅通过正常样本训练模型，缺乏异常样本，将会造成一定程度的误判问题，提出了一种基于有效异常样本构造的异常检测算法。通过K-means聚类算法得到代表不同类型正常事件的聚类簇，然后，基于异常事件的时序关系构造异常样本，再结合本文构造的异常样本，利用二分类支持向量机算法训练分类器，将检测任务转化为分类任务，从而提高检测准确率。本文在经典数据集（Avenue数据集）上进行了算法有效性验证，发现本文算法的检测准确度优于一些领域内的先进算法。因此，充分利用视频的时序关系进行异常样本的构造能有效提高异常检测的有效性。

关键词: 通信与信息系统, 视频处理, 异常检测, 支持向量机

Abstract:

At present， most anomaly detection algorithms only train models through normal samples， but the lack of abnormal sample may cause misjudgment. This paper proposes an anomaly detection algorithm based on effective anomaly sample construction. The clustering clusters representing different types of normal events are obtained by K-means clustering algorithm. Then， the abnormal samples are constructed based on the temporal relationship of abnormal events. Combined with the abnormal samples constructed in this paper， the classifier is trained by binary classification support vector machine algorithm to transform the detection task into classification task， thereby improve the detection accuracy. The effectiveness of the algorithm is verified on the classic dataset（the Avenue dataset）， and it is found that the detection accuracy of this algorithm is better than some advanced algorithms. Therefore， making full use of the timing relationship of video to construct abnormal samples can effectively improve the effectiveness of abnormal detection.

Key words: communication and information system, video processing, abnormality detection, support vector machines

中图分类号:

TN911.73

侯春萍,赵春月,王致芃,田海瑞. 基于有效异常样本构造的视频异常检测算法[J]. 吉林大学学报(工学版), 2021, 51(5): 1823-1829.

Chun-ping HOU,Chun-yue ZHAO,Zhi-peng WANG,Hai-rui TIAN. Video anomaly detection algorithm based on effective anomaly sample construction[J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(5): 1823-1829.

图/表 8

图1

图2

图3

图4

表1

图5

图6

表2

参考文献 19

1	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, Long Beach, 2019: 7842-7851.
2	孙静怡, 牟若瑾, 苏晓波. 基于支持向量机的高速公路异常状态检测[J]. 重庆交通大学学报: 自然科学版, 2019, 38(5): 1-7.
	Sun Jing-yi, Mu Ruo-jin, Su Xiao-bo. Anomaly detection of highway based on support vector machine[J]. Journal of Chongqing Jiaotong University(Natural Science Edition), 2019, 38(5): 1-7.
3	胡宏宇, 王庆年, 曲昭伟, 等. 运动目标空间模式辨识与异常交通行为检测[J]. 吉林大学学报: 工学版, 2011, 41(6): 598-602.
	Hu Hong-yu, Wang Qing-nian, Qu Zhao-wei, et al. Spatial pattern recognition of moving targets and abnormal traffic behavior detection[J]. Journal of Jilin University(Engineering and Technology Edition), 2011, 41(6): 598-602.
4	王洪雁, 周梦星. 基于光流及轨迹的人群异常行为检测[J]. 吉林大学学报: 工学版, 2020, 50(6): 2229-2237.
	Wang Hong-yan, Zhou Meng-xing. Detection of abnormal crowd behavior based on optical flow and trajectory[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(6): 2229-2237.
5	Cheng K W, Chen Y T, Fang W H. Video anomaly detection and localization using hierarchical feature representation and gaussian process regression[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, 2015: 2909-2917.
6	Cong Y, Yuan J, Liu J. Sparse reconstruction cost for abnormal event detection[C]∥Conference on Computer Vision and Pattern Recognition, Providence, 2011: 3449-3456.
7	Dutta J K, Banerjee B. Online detection of abnormal events using incremental coding length[C]∥Twenty-ninth Aaai Conference on Artiﬁcial Intelligence, Hawaii, 2015: 3755-3761.
8	Ionescu J R T, Smeureanu S, Popescu M, al et, Detecting abnormal events in video using narrowed normality clusters[C]∥IEEE Winter Conference on Applications of Computer Vision, Waikoloa Village, 2019: 1951-1960.
9	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.
10	胡凤苓, 孙少超. 智能视频监控中的群体异常事件检测研究[J]. 公安海警学院学报, 2015(4): 75-77.
	Hu Feng-ling, Sun Shao-chao. Research on group anomaly detection in intelligent video surveillance[J]. Journal of Public Security Maritime Police College, 2015(4): 75-77.
11	Fernando B, Bilen H, Gavves E, et al. Self-supervised video representation learning with odd-one-out networks[C]∥IEEE International Conference on Computer Vision and Pattern Recognition, Honolulu, 2017: 3636-3645.
12	Dwibedi D, Aytar Y, Tompson J, et al. Temporal cycle-consistency learning[C]∥Conference on Computer Vision and Pattern Recognition, Long Beach, 2019: 1801-1810.
13	Wang X, Jabri A, Efros A A. Learning correspondence from the cycle-consistency of time[C]∥Conference on Computer Vision and Pattern Recognition, Long Beach, 2019: 2561-2571.
14	Liu W, Luo W, Lian D, et al. Future frame prediction for anomaly detection—a new baseline[C]∥Conference on Computer Vision and Pattern Recognition, Salt Lake City, 2018: 6536-6545.
15	Hasan M, Choi J, Neumann J, et al. Learning temporal regularity in video sequences[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 2016: 733-742.
16	Vedaldi A, Fulkerson B. Vlfeat: an open and portable library of computer vision algorithms[C]∥Proceedings of the 18th ACM International Conference on Multimedia, Firenze, 2010: 1469-1472.
17	Du Q, Faber V, Gunzburger M. Centroidal voronoi tessellations: applications and algorithms[J]. SIAM Review, 1999, 41(4): 637-676.
18	Arthur D, Vassilvitskii S. K-means++: the advantages of careful seeding[C]∥Eighteenth Acm-Siam Symposium on Discrete Algorithms, New Orleans, 2007: 1027-1035.
19	Del Giorno A, Bagnell J, Hebert M. A discriminative framework for anomaly detection in large videos[C]∥European Conference on Computer Vision, Amsterdam, 2016: 334-349.

相关文章 15

[1]	院老虎,连冬杉,张亮,刘义. 基于密集连接卷积网络和支持向量机的飞行器机械部件故障诊断[J]. 吉林大学学报(工学版), 2021, 51(5): 1635-1641.
[2]	王康,姚猛,李立犇,李建桥,邓湘金,邹猛,薛龙. 基于月面表取采样触月压痕的月壤力学状态分析[J]. 吉林大学学报(工学版), 2021, 51(3): 1146-1152.
[3]	王义君,张有旭,缪瑞新,豆佳敏. 5G中基于系统中断概率的D2D资源分配算法[J]. 吉林大学学报(工学版), 2021, 51(1): 331-339.
[4]	胡钊政,李招康,陶倩文. 基于分布式二维激光测距仪的室内行人检测与跟踪[J]. 吉林大学学报(工学版), 2020, 50(2): 719-729.
[5]	李雄飞,王婧,张小利,范铁虎. 基于SVM和窗口梯度的多焦距图像融合方法[J]. 吉林大学学报(工学版), 2020, 50(1): 227-236.
[6]	谷远利, 张源, 芮小平, 陆文琦, 李萌, 王硕. 基于免疫算法优化LSSVM的短时交通流预测[J]. 吉林大学学报(工学版), 2019, 49(6): 1852-1857.
[7]	赵宏伟,李明昭,刘静,胡黄水,王丹,臧雪柏. 基于自然性和视觉特征通道的场景分类[J]. 吉林大学学报(工学版), 2019, 49(5): 1668-1675.
[8]	刘玉梅,乔宁国,庄娇娇,刘鹏程,胡婷,陈立军. 基于多传感器数据融合的轨道车辆齿轮箱异常检测[J]. 吉林大学学报(工学版), 2019, 49(5): 1465-1470.
[9]	卢洋,王世刚,赵文婷,赵岩. 基于离散Shearlet类别可分性测度的人脸表情识别方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1715-1725.
[10]	赵鹏,蒋宇中,陈斌,李春腾,张杨勇. 基于局部方差域自适应Blanking的超低频信道噪声抑制方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1696-1705.
[11]	赵金钢,张明,占玉林,谢明志. 基于塑性应变能密度的钢筋混凝土墩柱损伤准则[J]. 吉林大学学报(工学版), 2019, 49(4): 1124-1133.
[12]	陈俊,张奇峰,张艾群,蔡笃思. 基于深渊鱼类识别的原位自主观测方法[J]. 吉林大学学报(工学版), 2019, 49(3): 953-962.
[13]	隗海林, 包翠竹, 李洪雪, 李明达. 基于最小二乘支持向量机的怠速时间预测[J]. 吉林大学学报(工学版), 2018, 48(5): 1360-1365.
[14]	耿庆田, 于繁华, 王宇婷, 高琦坤. 基于特征融合的车型检测新算法[J]. 吉林大学学报(工学版), 2018, 48(3): 929-935.
[15]	蔡振闹, 吕信恩, 陈慧灵. 基于反向细菌优化支持向量机的躯体化障碍预测模型[J]. 吉林大学学报(工学版), 2018, 48(3): 936-942.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

d	AUC
0.9	0.786
1.0	0.790
1.1	0.794
1.2	0.798
1.3	0.795
1.4	0.794

算法	AUC
文献［15］	0.702
文献［19］	0.783
文献［8］	0.788
K-means+one-class SVM	0.576
本文	0.798