Journal of Shandong University(Engineering Science) ›› 2018, Vol. 48 ›› Issue (6): 56-66.doi: 10.6040/j.issn.1672-3961.0.2018.202

• Machine Learning & Data Mining • Previous Articles     Next Articles

Multi-node human behavior recognition based on linear acceleration

Xing LI1(),Zhenjie HOU1,Jiuzhen LIANG1,Xingzhi CHANG2   

  1. 1. College of Information Science and Engineering, Changzhou University, Changzhou 213164, Jiangsu, China
    2. Changzhou Key Laboratory of Large Plastic Parts Intelligence Manufacturing, Changzhou College of Information Technology, Changzhou 213164, Jiangsu, China
  • Received:2018-05-25 Online:2018-12-20 Published:2018-12-26
  • Supported by:
    国家自然科学基金项目(61063021);江苏省产学研前瞻性联合研究项目(BY2015027-12);江苏省物联网移动互联技术工程重点实验室开放课题项目(JSWLW-2017-013)

RichHTML

9

PDF (PC)

723

Abstract:

Focused on the issue that the behavior data in the current acceleration-based human behavior recognition method was affected by the gravitational acceleration and the lack of spatial information, a multi-node human behavior recognition algorithm based on linear acceleration was proposed. The linear acceleration was obtained by removing gravitational acceleration using segmented bi-directionally removal of gravitational acceleration algorithm. The tremor motion signal was filtered by a sliding averaging filter for linear acceleration and sensor acceleration, and the redundant actions in the two accelerations were cropped. The dynamic time warping (DTW) distance features between different joint points and seven conventional time domain features were extracted from two accelerations. The support vector machine was employed to recognize the human behavior. Experimental results showed that this method could effectively improve the accuracy of human behavior recognition.

Key words: multi-node, linear acceleration, DTW distance feature, support vector machine

CLC Number: 

  • TP391

Fig.1

Diagram of behaviors"

Fig.2

The flow chart of our algorithm"

Fig.3

The algorithm of segmental bi-directional removal of gravitational acceleration"

Fig.4

The angle signals of three axises when stationary"

Fig.5

The comparison of sensor acceleration before and after filtering"

Fig.6

The comparison of linear acceleration before and after filtering"

Fig.7

The linear acceleration of raising hand"

Fig.8

The linear acceleration of putting down hand"

Table 1

Initial feature set"

加速度类型 单节点特征 两两关节点间特征
手腕 肘部
传感器加速度 7特征 7特征 DTW距离特征
线性加速度 7特征 7特征 DTW距离特征

Fig.9

The recognition rate of clipping redundant action underdifferent parameter combinations"

Table 2

The recognition rate and operation efficiency of two algorithms"

方法类别 识别率/% 提取时间/s 训练时间/s 测试时间/s 总的时间/s
原始方法 85.33 0.002 1 0.008 0 0.004 6 0.014 7
本研究方法 98.89 0.004 5 0.008 0 0.003 9 0.016 4

Table 3

Optimization results of the feature set"

加速度 单节点特征 两两关节点间特征
手腕 肘部
传感器加速度 2、5、6 1~7 8~10
线性加速度 1~7 1、2、4~6 8~10

Table 4

Experiment results of A1 method"

基础方法 识别率/%
添加A1方法前 添加A1方法后
A0 85.33 93.78
A0+A2 85.78 94.67
A0+A2+A3 88.44 95.33
A0+A2+A3+A5 92.44 97.78

Table 5

Experiment results of A2 method"

基础方法 识别率/%
添加A2方法前 添加A2方法后
A0 85.33 85.78
A0+A1 93.78 94.67
A0+A1+A3 94.44 95.33
A0+A1+A3+A4 94.89 95.56
A0+A1+A3+A4+A5 97.56 98.89

Table 6

Experiment results of A3 method"

基础方法 识别率/%
添加A3方法前 添加A3方法后
A0 85.33 86.89
A0+A1 93.78 94.44
A0+A1+A2 94.67 95.33
A0+A1+A2+A4 95.11 95.56
A0+A1+A2+A4+A5 98.22 98.89

Table 7

Experiment results of A4 method"

基础方法 识别率/%
添加A4方法前 添加A4方法后
A0+A1 93.78 94.44
A0+A1+A2 94.67 95.11
A0+A1+A2+A3 95.33 95.56
A0+A1+A2+A3+A5 97.78 98.89

Table 8

Experiment results of A5 method"

基础方法 识别率/%
添加A5方法前 添加A5方法后
A0 85.33 91.33
A0+A1 93.78 95.56
A0+A1+A2 94.67 97.56
A0+A1+A2+A3 95.33 97.78
A0+A1+A2+A3+A4 95.56 98.89

Fig.10

Confusion matrix"

Table 9

Comparison experiment result of our method andanother method"

方法 识别率/%
原始方法 85.33
文献[5]方法 94.44
本研究方法 98.89
1 MUKHOPADHYAY S C . Wearable sensors for human activity monitoring: a review[J]. IEEE Sensors Journal, 2014, 15 (3): 1321- 1330.
2 EDWARDS J . Wireless sensors relay medical insight to patients and caregivers:special reports[J]. IEEE Signal Processing Magazine, 2012, 29 (3): 8- 12.
doi: 10.1109/MSP.2012.2183489
3 KHAN A M , LEE Y K , LEE S Y , et al. A triaxial acclerometer-based physical-activity recognition via augmented-signal features and a hierarchical recognizer[J]. IEEE Transactions on Information Technology in Biomedicine, 2010, 14 (5): 1166- 1172.
4 LEE S W , MASE K . Activity and location recognition using wearable sensors[J]. Pervasive Computing IEEE, 2002, 1 (3): 24- 32.
5 CHEN Y, XUE Y. A deep Learning approach to human activity recognition based on single accelerometer[C]//IEEE International Conference on System, Man, and Cybernetics. Budapest, Hungary: IEEE, 2016: 1488-1492.
6 BRZOSTOWSKI K , DRAPALA J , GRZECH A , et al. Adaptive decision support systems for automatic physical effort plan generation: data-driven approach[J]. Journal of Cybernetics, 2013, 44 (2-3): 204- 221.
7 池小文, 倪友聪, 杜欣, 等. 基于智能手机的健身跑数据采集及演化建模[J]. 智能系统学报, 2017, 5 (12): 702- 709.
CHI Xiaowei , NI Youcong , DU Xin , et al. Smartphone-based speed acquisition and evolutionary modeling for fitness running[J]. CAAI Transactions on Intelligent Systems, 2017, 5 (12): 702- 709.
8 REHBINDER H , HU X . Drift free attitude estimation for accelerated rigid bodies[J]. Automatica, 2004, 40 (4): 653- 659.
doi: 10.1016/j.automatica.2003.11.002
9 李文锋, 姚丙萌. 基于单三轴加速度传感器的人体活动状态识别[J]. 华中科技大学学报(自然科学版), 2016, 44 (4): 58- 62.
LI Wenfeng , YAO Bingmeng . Recognition of human activities based on single triaxial accelerometer[J]. Journal of Huazhong University of Science and Technology(Natural Science Edition), 2016, 44 (4): 58- 62.
10 王昌喜.基于加速度信息的上肢动作识别系统设计及动作质量评价方法的研究[D].合肥:中国科学技术大学, 2010.
WANG Changxi. Design of recongnition system and study of quality evaluation for upper limb movement based on acceleration[D]. Hefei: University of Science and Technology of China, 2010.
11 JOSHUA L , VARGHESE K . Accelerometer-based activity recognition in construction[J]. Journal of Computing in Civil Engineering, 2011, 25 (5): 370- 379.
doi: 10.1061/(ASCE)CP.1943-5487.0000097
12 JOSHUA L , VARGHESE K . Automated recognition of construction labour activity using accelerometers in field situations[J]. International Journal of Productivity and Performance Management, 2014, 63 (7): 841- 862.
13 SAKOE H . Two-level DP-matching:a dynamic programming-based pattern matching algorithm for connected word recognition[J]. Readings in Speech Recognition, 1990, 27 (6): 180- 187.
14 SAENPHON T , PHIMOLTARES S , LURSINSAP C . Combining new fast opposite gradient search with ant colony optimization for solving travelling salesman problem[J]. Engineering Applications of Artificial Intelligence, 2014, 35 (2): 324- 334.
15 ESCARIO J B , JIMENEZ J F , GIRON-SIERRA J M . Ant colony extended: experiments on the travelling salesman problem[J]. Expert Systems with Applications, 2015, 42 (1): 390- 410.
doi: 10.1016/j.eswa.2014.07.054
16 凌佩佩, 邱崧, 蔡茗名, 等. 结合特权信息的人体动作识别[J]. 中国图像图形学报, 2017, 22 (4): 482- 491.
LING Peipei , QIU Song , CAI Mingming , et al. Human action recognition based on privileged information[J]. Journal of Image and Graphics, 2017, 22 (4): 482- 491.
17 HOU C , NIE F , ZHANG C , et al. Multiple rank multi-linear SVM for matrix data classification[J]. Pattern Recognition, 2014, 47 (1): 454- 469.
18 IOSIFIDIS A , GABBOUJ M . Multi-class support vector machine classifiers using intrinsic and penalty graphs[J]. Pattern Recognition, 2016, 55, 231- 246.
doi: 10.1016/j.patcog.2016.02.002
19 叶明全, 高凌云, 万春圆. 基于人工蜂群和SVM的基因表达数据分类[J]. 山东大学学报(工学版), 2018, 48 (3): 10- 16.
YE Mingquan , GAO lingyun , WAN Chunyuan . Gene expression data classification based on artificial bee colony and SVM[J]. Journal of Shandong University(Engineering Science), 2018, 48 (3): 10- 16.
20 CHEN Chen , ZHANG Baochang , HOU Zhenjie , et al. Action recognition from depth sequences using weighted fusion of 2D and 3D auto-correlation of gradients features[J]. Multimedia Tools and Applications, 2017, 76 (3): 4651- 4669.
[1] Xin MA,Xue WANG. Prediction of microRNA-binding residues based on Laplacian support vector machine and sequence information [J]. Journal of Shandong University(Engineering Science), 2020, 50(2): 76-82.
[2] Zhixiang LIANG,Xiaoming LIU,Ying MU,Yutian LIU. Prediction method of wind power and PV ramp event based on deep learning [J]. Journal of Shandong University(Engineering Science), 2019, 49(5): 24-28.
[3] YE Mingquan, GAO Lingyun, WAN Chunyuan. Gene expression data classification based on artificial bee colony and SVM [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2018, 48(3): 10-16.
[4] HAN Xueshan, WANG Junxiong, SUN Donglei, LI Wenbo, ZHANG Xinyi, WEI Zhiqing. Nodal load forecasting method considering spatial correlation and redundancy [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(6): 7-12.
[5] LI Sushu, WANG Shitong, LI Tao. A feature selection method based on LS-SVM and fuzzy supplementary criterion [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(3): 34-42.
[6] LIU Jie, YANG Peng, LYU Wensheng, LIU Agudamu, LIU Junxiu. Prediction models of PM2.5 mass concentration based on meteorological factors [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2015, 45(6): 76-83.
[7] LIU Xiaoyong. A semi-supervised method based on tree kernel for relationship extraction [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2015, 45(2): 22-26.
[8] HAO Qingbo, MU Shaomin, YIN Chuanhuan, CHANG Tengteng, CUI Wenbin. An algorithm of fast local support vector machine based on clustering [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2015, 45(1): 13-18.
[9] LI Faquan, YANG Licai, YAN Hongbo. An emotion recognition method of multiphysiological information fusion based on PCA-SVM [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2014, 44(6): 70-76.
[10] WANG Xiao-feng, SUI Ting-ting. Protein sequence identification based on improved TIGA-S4VM algorithm [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2014, 44(1): 1-6.
[11] WANG Hao, HUA Ji-xue, FAN Xiao-shi. Intrusion detection technology based on twin support vector machine [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2013, 43(6): 53-56.
[12] LI Fu-gui1,2, HUANG Tian-qiang1,2*, SU Li-chao1,2, SU Wei-feng3. Heterologous video copy-move forgery detection by fusing multiple features [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2013, 43(4): 32-38.
[13] SHI Jun, ZHU Min. An optimization model for forecasting based on grey system and support vector machine [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2012, 42(5): 7-11.
[14] ZHAO Jia-min, FENG Ai-min*, LIU Xue-jun. A new structured one-class support vector machine with local density embedding [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2012, 42(4): 13-18.
[15] PAN Dong-yin, ZHU Fa, XU Sheng, YE Ning*. Feature selection of gene expression profiles of colon cancer [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2012, 42(2): 23-29.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] WANG Li-ju,HUANG Qi-cheng,WANG Zhao-xu . [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2006, 36(6): 51 -56 .
[2] WANG Jing,LI Yu-jiang,ZHANG Xiao-jin,BI Yan-jun,CHEN Wei-suo . [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2006, 36(6): 100 -103 .
[3] XIA Bin,ZHANG Lian-jun . Energy comparison-based TOA estimation algorithm for the DS-CDMA UWB system[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2007, 37(1): 70 -73 .
[4] CHEN Huaxin, CHEN Shuanfa, WANG Binggang. The aging behavior and mechanism of base asphalts[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2009, 39(2): 125 -130 .
[5] LIANG Jing-yun,WANG Ming-gang,CHAI Jia-qian,LIU yong-qing . Synthesis and in vitro antibacterial activity of 1,6-Di-(N5-phenyl-N1-diguanido) hexane dihydrochloride[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(3): 104 -107 .
[6] ZHANG Gong-xiao,YANG Rong-hua . Synthesis and characterization of salicylaldehyde methylthiosemicarbazone Schiff base complexes[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(3): 108 -111 .
[7] XUE Qiang,AI Xing,ZHAO Jun,ZHOU Yong-hui,YUAN Xun-liang . Effects of TiC nano-sized particle on the microstructure and properties of Si3N4 composite ceramics[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(3): 69 -72 .
[8] ZHANG Li,LI Qing-min,LI Qing-quan . Trends of EMC problem generated by FACTS equipment[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2007, 37(2): 52 -57 .
[9] DIAO Yong, TIAN Si-Meng, CAO Zhe-Meng. Geological work method for the construction of the Yichang Wanzhou Railway tunnel in high risk karst areas[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2009, 39(5): 91 -95 .
[10] SONG Qing,LI Xiao-lei,ZHANG Cheng-jin . Optimization of a postal express mail network based on bottleneck analysis[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2007, 37(5): 29 -33 .
Copyright 2018 © Editorial office of Journal of Shandong University (Engineering Science)
Supported by Beijing Magtech Co.Ltd