基于人体关节点数据的攻击性行为识别

doi:10.11772/j.issn.1001-9081.2019010084

计算机应用 ›› 2019, Vol. 39 ›› Issue (8): 2235-2241.DOI: 10.11772/j.issn.1001-9081.2019010084

基于人体关节点数据的攻击性行为识别

陈皓, 肖利雪, 李广, 潘跃凯, 夏雨

西安邮电大学计算机学院, 西安 710121

收稿日期:2019-01-15 修回日期:2019-03-08 发布日期:2019-04-15 出版日期:2019-08-10
通讯作者: 陈皓
作者简介:陈皓(1978-),男,河北安新人,副教授,博士,CCF会员,主要研究方向:进化计算、工程优化;肖利雪(1992-),女,内蒙古赤峰人,硕士研究生,主要研究方向:计算机智能、数据挖掘;李广(1995-),男,陕西铜川人,硕士研究生,主要研究方向:计算机智能、数据挖掘;潘跃凯(1995-),男,山东聊城人,硕士研究生,主要研究方向:计算机智能、数据挖掘;夏雨(1996-),女,陕西咸阳人,硕士研究生,主要研究方向:计算机智能、数据挖掘。
基金资助:
国家自然科学基金资助项目（61203311，61105064）；陕西省教育厅自然科学专项（17JK0701）；陕西省网络数据分析与智能处理重点实验室开放课题基金资助项目（XUPT-KLND （201804））；西安邮电大学创新基金资助项目（CXJJ2017069）。

Aggressive behavior recognition based on human joint point data

CHEN Hao, XIAO Lixue, LI Guang, PAN Yuekai, XIA Yu

School of Computer Science & Technology, Xi'an University of Posts & Telecommunications, Xi'an Shaanxi 710121, China

Received:2019-01-15 Revised:2019-03-08 Online:2019-04-15 Published:2019-08-10
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61203311, 61105064), the Scientific Research Program Funded by Shaanxi Provincial Education Department of China (17JK0701), the Science Foundation of the Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing (XUPT-KLND (201804)), the Graduate Innovation Fund of Xi'an University of Posts & Telecommunications (CXJJ2017069).

摘要/Abstract

摘要： 针对人体攻击性行为识别问题，提出一种基于人体关节点数据的攻击性行为识别方法。首先，利用OpenPose获得单帧图像中的人体关节点数据，并通过最近邻帧特征加权法和分段多项式回归完成由人体自遮挡和环境因素所导致缺失值的补全；然后，对每个人体定义动态"安全距离"阈值，如果两人真实距离小于阈值，则构建行为特征矢量，其中包括帧间人体重心位移、人体关节旋转角角速度和发生交互时的最小攻击距离等；最后，提出改进的LightGBM算法w-LightGBM，并对攻击性行为进行识别。采用公共数据集UT-interaction对所提出的攻击性行为分类识别方法进行测试实验，准确率达到95.45%。实验结果表明，所提方法能够有效识别各种角度的攻击性行为。

关键词: 人体关节点数据, 攻击性行为识别, 缺失值补全, 攻击距离

Abstract: In order to solve the problem of human aggressive behavior recognition, an aggressive behavior recognition method based on human joint points was proposed. Firstly, OpenPose was used to obtain the human joint point data of a single frame image, and nearest neighbor frame feature weighting method and piecewise polynomial regression were used to realize the completion of missing values caused by body self-occlusion and environmental factors. Then, the dynamic "safe distance" threshold was defined for each human body. If the true distance between the two people was less than the threshold, the behavior feature vector was constructed, including the human barycenter displacement between frames, the angular velocity of human joint rotation and the minimum attack distance during interaction. Finally, the improved LightGBM (Light Gradient Boosting Machine) algorithm, namely w-LightGBM (weight LightGBM), was used to realize the classification and recognition of aggressive behaviors. The public dataset UT-interaction was used to verify the proposed method, and the accuracy reached 95.45%. The results show that this method can effectively identify the aggressive behaviors from various angles.

Key words: human joint point data, aggressive behavior recognition, missing value completion, attack distance

中图分类号:

TP18

陈皓, 肖利雪, 李广, 潘跃凯, 夏雨. 基于人体关节点数据的攻击性行为识别[J]. 计算机应用, 2019, 39(8): 2235-2241.

CHEN Hao, XIAO Lixue, LI Guang, PAN Yuekai, XIA Yu. Aggressive behavior recognition based on human joint point data[J]. Journal of Computer Applications, 2019, 39(8): 2235-2241.

参考文献

[1] SCHWAPPACH D, SENDLHOFER G, HÄSLER L, et al. Speaking up behaviors and safety climate in an Austrian university hospital[J]. International Journal for Quality in Health Care, 2018, 30(9):701-707.
[2] YU S, CHEN H, BROWN R A. Hidden Markov model-based fall detection with motion sensor orientation calibration:a case for real-life home monitoring[J]. IEEE Journal of Biomedical and Health Informatics, 2018, 22(6):1847-1853.
[3] RABIEE H, MOUSAVI H, NABI M, et al. Detection and localization of crowd behavior using a novel tracklet-based model[J]. International Journal of Machine Learning and Cybernetics, 2017, 9(12), 1999-2010.
[4] LU X D, KOGA T. Real-time motion detection for high-assurance aircraft tracking system using downlink aircraft parameters[J]. Simulation Modelling Practice and Theory, 2016, 65:81-92.
[5] STONE E E, SKUBIC M. Fall detection in homes of older adults using the Microsoft Kinect[J]. IEEE Journal of Biomedical and Health Informatics, 2015, 19(1):290-301.
[6] CHANG X, MA Z, LIN M, et al. Feature interaction augmented sparse learning for fast Kinect motion detection[J]. IEEE Transactions on Image Processing, 2017, 26(8):3911-3920.
[7] OUANANE A, SERIR A. New paradigm for recognition of aggressive human behavior based on bag-of-features and skeleton graph[C]//Proceedings of the 8th International Workshop on Systems, Signal Processing and Their Applications. Piscataway, NJ:IEEE, 2013:133-138.
[8] SERRANO I, DENIZ O, BUENO G, et al. Spatio-temporal elastic cuboid trajectories for efficient fight recognition using Hough forests[J]. Machine Vision & Applications, 2018, 29(2):207-217.
[9] CAO Z, SIMON T, WEI S. et al. Realtime multi-person 2D pose estimation using part affinity fields[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2017:1302-1310.
[10] MERAD D, AZIZ K, IGUERNAISSI R, et al. Tracking multiple persons under partial and global occlusions:application to customers' behavior analysis[J]. Pattern Recognition Letters, 2016, 81:11-20.
[11] RAZARI-FAR R, CHAKRABARTI S, SAIF M, et al. An integrated imputation-prediction scheme for prognostics of battery data with missing observations[J]. Expert Systems with Applications, 2016, 115:709-723.
[12] 史慧.谈谈人际交往中的空间距离[J].河南水利与南水北调,2007(7):73-74. (SHI H, Talking about the spatial distance in interpersonal communication[J]. Henan Water Resources and South-to-North Water Diversion, 2007(7):73-74.)
[13] 费章惠.中国大百科全书:力学[M].北京:中国大百科全书出版社,1987:133. (FEI Z H. Encyclopedia of China:Mechanics[M]. Beijing:Encyclopedia of China Publishing House, 1987:133.)
[14] MA X, SHA J, WANG D, et al. Study on a prediction of P2P network loan default based on the machine learning LightGBM and XGboost algorithms according to different high dimensional data cleaning[J]. Electronic Commerce Research and Applications, 2018, 31:24-39.
[15] CHEN W, FU K, ZUO J, et al. Radar emitter classification for large data set based on weighted-XGboost[J]. IET Radar, Sonar & Navigation, 2017, 11(8):1203-1207.
[16] SENER F, IKIZLER-CINBIS N. Two-person interaction recognition via spatial multiple instance embedding[J]. Journal of Visual Communication & Image Representation, 2015, 32:63-73.
[17] ZHANG B, ROTA P, CONCI N, et al. Human interaction recognition in the wild:analyzing trajectory clustering from multiple-instance-learning perspective[C]//Proceedings of the 2015 IEEE International Conference on Multimedia and Expo. Piscataway, NJ:IEEE, 2015:1-6.
[18] KONG Y, LIANG W, DONG Z, et al. Recognising human interaction from videos by a discriminative model[J]. IET Computer Vision, 2014, 8(4):277-286.
[19] KONG Y, FU Y. Modeling supporting regions for close human interaction recognition[C]//Proceedings of the 2014 European Conference on Computer Vision, LNCS 8926. Berlin:Springer, 2014:29-44.
[20] NGUYEN N, YOSHITAKA A. Human interaction recognition using independent subspace analysis algorithm[C]//Proceedings of the 2014 IEEE International Symposium on Multimedia.Piscataway, NJ:IEEE, 2014:40-46.
[21] el houda SLIMANI K N, BENEZETH Y, SOUAMI F. Human interaction recognition based on the co-occurrence of visual words[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Piscataway, NJ:IEEE, 2014:461-466.
[22] 王佩瑶,曹江涛,姬晓飞.基于改进时空兴趣点特征的双人交互行为识别[J].计算机应用,2016, 36(10):2875-2879. (WANG P Y, CAO J T, JI X F. Two-person interactive behavior recognition based on improved spatio-temporal interest points[J]. Journal of Computer Applications, 2016, 36(10):2875-2879.)
[23] 姬晓飞,左鑫孟.基于分阶段视觉共生矩阵序列的双人交互识别[J].计算机工程与设计,2017,38(9).2498-2503. (JI X F, ZUO X M. Human interaction recognition based on multi-stag framework and co-occurring visual matrix sequence[J]. Computer Engineering and Design, 2017, 38(9):2498-2503.)

基于人体关节点数据的攻击性行为识别

Aggressive behavior recognition based on human joint point data

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	宫智宇王士同. 面向重尾噪声图像分类的残差网络学习方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[2]	王虎王晓峰李可马云洁. 融合多头自注意力的标签语义嵌入联邦类增量学习方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[3]	丁建立, 黄辉, 曹卫东. 航班链运行状态动态监控方法[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3941-3948.
[4]	刘晶鑫, 黄雯静, 徐亮胜, 黄冲, 吴建生. 字典学习与样本关联保持结合的无监督特征选择模型[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3766-3775.
[5]	宋逸飞, 柳毅. 基于数据增强和标签噪声的快速对抗训练方法[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3798-3807.
[6]	沈嫣然, 温昕, 张瑾昊, 张帅, 曹锐, 高保禄. 轻量级多尺度卷积网络的功能磁共振成像脑龄预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3949-3957.
[7]	张祖篡, 陈学斌, 高瑞, 邹元怀. 基于标签分类的联邦学习客户端选择方法[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3759-3765.
[8]	蒋权黄文清苟志勇. 基于等变图神经网络的拉格朗日粒子流模拟[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[9]	李岚皓严皓钧周号益孙庆赟李建欣. 基于神经网络的多尺度信息融合时间序列长期预测模型[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[10]	廖炎华鄢元霞潘文林. 基于YOLOv9的交通路口图像的多目标检测算法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[11]	张学飞张丽萍闫盛侯敏赵宇博. 知识图谱与大语言模型协同的个性化学习推荐[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[12]	索晋贤张丽萍闫盛王东奇张雅雯. 可解释的深度知识追踪方法综述[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[13]	陈丹阳张长伦. 多尺度去相关的图卷积网络[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[14]	蒋沛宇王永光任亚亭李硕晨谭火彬. 基于测量不确定度表示指南的目标检测不确定度测量方案[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[15]	赵小阳许新征李仲年. 物联网应用中的可解释人工智能研究综述[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.