[1]赵 雄,陈 平,潘晋孝.基于骨架关键点的车内异常行为识别方法[J].机械与电子,2021,(03):10-15.
　ZHAO Xiong,CHEN Ping,PAN Jinxiao.Recognition Method of Abnormal Behavior in Car Based on Skeleton Key Points[J].Machinery & Electronics,2021,(03):10-15.
点击复制

基于骨架关键点的车内异常行为识别方法()
分享到：

机械与电子[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2021年03期
页码:
10-15
栏目:
设计与研究
出版日期:
2021-03-24

文章信息/Info

Title:
Recognition Method of Abnormal Behavior in Car Based on Skeleton Key Points
文章编号:
1001-2257(2021)03-0010-06
作者:
赵 雄; 陈 平; 潘晋孝
中北大学信息探测与处理山西省重点实验室,山西 太原 030051
Author(s):
ZHAO Xiong; CHEN Ping; PAN Jinxiao
Shanxi Key Laboratory of Signal Capturing and Processing, North University of China,Taiyuan 030051, China
关键词:
lpha pose; 坐姿骨架关键点; 概率学习模型
Keywords:
Alpha pose;  key points of sitting skeleton;  probabilistic learning model;  abnormal behavior recognition in the car
分类号:
TP391.4
文献标志码:
A
摘要:
针对现有异常行为识别方法在车内场景应用少，并且受车内空间狭小、异常行为复杂多变等影响导致识别有效性差等问题。在Alpha pose模型提取驾乘人员骨架关键点基础上，构建驾乘人员人体坐姿模型，采用关键点位置信息描述异常状态，最后利用概率学习模型将位置信息转换为概率对行为进行识别分类。经实验测试，该方法对车内前排人员异常行为的识别准确率能够达到90%以上，且具有一定的实用价值。
Abstract:
In view of the fact that the existing abnormal behavior identification methods is rarely used in the car scene, and due to the small space in the car, abnormal behaviors are complex and changeable, resulting in poor recognition effectiveness and other problems, the Alpha Pose model was used to extract the key points of the driver’s skeleton, the sitting posture model of the driver’s body was constructed, and the position information of the key points was used to describe the abnormal state.Finally, the probabilistic learning model is used to transform location information into probability to identify and classify behaviors. The experimental results show that the recognition accuracy of the method can reach more than 90%, and it has a certain practical value.

参考文献/References:

[1]COLLINS R T, LIPTON A J, KANADE T.Introduction to the special section on video surveillance[J]. IEEE Transactions on pattern analysis and machine intelligence, 2000, 22(8): 745-746.

[2]CONG Y, YUAN J S, LIU J. Sparse reconstruction cost for abnormal event detection[C]//2011 IEEE Conference on Computer Vision and Pattern Recognition, 2011:3449-3456.

[3]张进,李伟,俞建成,等.基于组合分类器的不同状态下脑电信号分类[J].控制与决策,2019,34(5):897-907.

[4]李文锋,余志刚,胡心韵.肌电信号选择对下肢关节连续运动估计的影响[J].机械设计与制造,2019(3):54-57.

[5]李江天,李敏,宋战兵.基于多源生理信号的驾驶疲劳检测[J].物流技术,2018,37(2):78-83.

[6]王至恒.基于车辆运行参数的驾驶危险状态辨识仿真研究[D].西安：长安大学,2018.

[7]DALAL N, TRIGGS B, SCHMID C.Human detection using oriented histograms of flow and appearance[C]// European Conference on Computer Vision.Berlin Heidelberg:Springer,2006:428-441.

[8]李此君,刘云鹏.基于协方差流形和LogitBoost的异常驾驶行为识别方法[J].激光与光电子学进展,2018, 55(11):332-339.

[9]衡红军,刘静.基于混合方法的多维时间序列驾驶异常点检测[J].计算机工程,2020,46(3):99-104.

[10]SODEMANN A A , ROSS M P , BORGHETTI B J.A review of anomaly detection in automated surveillance[J]. IEEE Transactions on systems, man, and cybernetics:part C,2012,42(6):1257-1272.

[11]HAMALAINEN W , NYKANEN M . Efficient discovery of statistically significant association rules[C]//Eighth IEEE International Conference on Data Mining. New York:IEEE,2008:203-212.

[12]AMRAEE S,VAFAEI A,JAMSHIDI K, et al. Anomaly detection and localization in crowded scenes using connected component analysis[J].Multimedia tools and applications, 2018,77(12):14767-14782.

[13]YUAN Y,FENG Y,LU X.Statistical hypothesis detector for abnormal event detection in crowded scenes[J]. IEEE Transactions on cybernetics,2017, 47(11):3597-3608.

[14]王鑫,侯志强,余旺盛,等.基于多层卷积特征融合的目标尺度自适应稳健跟踪[J].光学学报,2017,37(11):224-235.

[15]ZHOU S F,SHEN W,ZENG D,et al.Unusual event detection in crowded scenes by trajectory analysis[C]// 2015 IEEE International Conference on Acoustics, Speech and Signal Processing. New York:IEEE, 2015:1300-1304.

[16]周培培,丁庆海,罗海波,等.基于DBSCAN聚类算法的异常轨迹检测[J]. 红外与激光工程, 2017, 46(5):230-237.

[17]TRAN D,YUAN J S,FORSYTH D.Video Event detection: from subvolume localization to spatiotemporal path search[J]. IEEE Transactions on pattern analysis and machine intelligence, 2014, 36(2):404-416.

[18]NGUYEN N T,PHUNG D Q,VENKATESH S,et al.Learning and detecting activities from movement trajectories using the hierarchical hidden Markov model[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition.New York:IEEE, 2005:955-960.

[19]LEE D G,SUK H I,LEE S W.Crowd behavior representation using motion influence matrix for anomaly detection[C]//2013 2nd IAPR Asian Conference on Pattern Recognition.New York: IEEE, 2013:110-114.

[20]MEHRAN R, OYAMA A, SHAH M.Abnormal crowd behavior detection using social force model[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition. New York:IEEE, 2009:935-942.

[21]张鑫.滴滴出行与广东公安合作 “一键报警”展开深度警企联动[EB/OL]. (2019-1-10) https://www.sohu.com/a/287972021_255783.

[22]IQBAL U,MILAN A,GALL J.PoseTrack: joint multi-person pose estimation and tracking[C]// IEEE Conference on Computer Vision and Pattern Recognition.New York:IEEE, 2017:4654-4663.

[23]吴垒,于哲舟.基于概率神经网络的在线分类器[J].计算机技术与发展, 2019,29(12):14-20.

[24]王亚平,吴夷杉.概率神经网络在下肢运动趋势判断中的应用[J].机械设计与制造,2019(2):22-25.

[25]郑亚莉,王康.基于概率神经网络的垃圾邮件分类[J].计算机与现代化, 2008(1):8-10.

备注/Memo

备注/Memo:
收稿日期：2020-09-24
基金项目：山西省研究生创新项目资助 (2020BY098)；国 家 自 然 科 学 基 金 资 助 项 目(61801437,61871351,61971381)；山 西 省 自 然 科 学 基 金 资 助 项 目 (201801D221206,201801D221207)
作者简介：赵 雄（1994—），男，山西运城人，硕士研究生，研究方向为大数据技术、深度学习等，通信作者；陈 平（1983—）男，山西太原人，教授，博士，研究方向为信号与信息处理、图像处理与重建、人工智能等；潘晋孝（1965—）男，山西太原人，教授，博士，研究方向为矩阵理论、现代优化理论、图像信息处理及增强等。
更新日期/Last Update: 2021-03-22