Compressed sensing method for human activity recognition using tri-axis accelerometer on mobile phone

doi:10.1016/S1005-8885(17)60196-1

中国邮电高校学报(英文) ›› 2017, Vol. 24 ›› Issue (2): 31-37.doi: 10.1016/S1005-8885(17)60196-1

• Artificial Intelligence • 上一篇下一篇

Compressed sensing method for human activity recognition using tri-axis accelerometer on mobile phone

宋辉,王忠民

西安邮电大学

收稿日期:2016-09-19 修回日期:2017-04-06 出版日期:2017-04-30 发布日期:2017-04-30
通讯作者: 宋辉 E-mail:shnew@163.com
基金资助:
中国国家自然科学基金

Compressed sensing method for human activity recognition using tri-axis accelerometer on mobile phone

Hui SONG,Zhong-Min WANG

Received:2016-09-19 Revised:2017-04-06 Online:2017-04-30 Published:2017-04-30
Contact: Hui SONG E-mail:shnew@163.com
Supported by:
the National Natural Science Foundation of China (61373116).

摘要/Abstract

摘要： The diversity in the phone placements of different mobile users’ dailylife increases the difficulty of recognizing human activities by using mobile phone accelerometer data. To solve this problem, a compressed sensing method to recognize human activities that is based on compressed sensing theory and utilizes both raw mobile phone accelerometer data and phone placement information is proposed. First, an over-complete dictionary matrix is constructed using sufficient raw tri-axis acceleration data labeled with phone placement information. Then, the sparse coefficient is evaluated for the samples that need to be tested by resolving L1 minimization. Finally, residual values are calculated and the minimum value is selected as the indicator to obtain the recognition results. Experimental results show that this method can achieve a recognition accuracy reaching 89.86%, which is higher than that of a recognition method that does not adopt the phone placement information for the recognition process. The recognition accuracy of the proposed method is effective and satisfactory.

关键词: activity recognition, compressed sensing, mobile phone accelerometer, phone placements

Abstract: The diversity in the phone placements of different mobile users’ dailylife increases the difficulty of recognizing human activities by using mobile phone accelerometer data. To solve this problem, a compressed sensing method to recognize human activities that is based on compressed sensing theory and utilizes both raw mobile phone accelerometer data and phone placement information is proposed. First, an over-complete dictionary matrix is constructed using sufficient raw tri-axis acceleration data labeled with phone placement information. Then, the sparse coefficient is evaluated for the samples that need to be tested by resolving L1 minimization. Finally, residual values are calculated and the minimum value is selected as the indicator to obtain the recognition results. Experimental results show that this method can achieve a recognition accuracy reaching 89.86%, which is higher than that of a recognition method that does not adopt the phone placement information for the recognition process. The recognition accuracy of the proposed method is effective and satisfactory.

Key words: activity recognition, compressed sensing, mobile phone accelerometer, phone placements

Hui SONG Zhong-Min WANG. Compressed sensing method for human activity recognition using tri-axis accelerometer on mobile phone[J]. JOURNAL OF CHINA UNIVERSITIES OF POSTS AND TELECOM, 2017, 24(2): 31-37.

参考文献

1. Shoaib M, Bosch S, Durmaz Incel O D, et al. A survey of online activity recognition using mobile phones. Sensors, 2015, 15(1): 2059-2085

2. Su X, Tong H H, Ji P. Activity recognition with smartphone sensors. Tsinghua Science and Technology, 2014, 19(3): 235-249

3. Incel O D, Kose M, Ersoy C. A review and taxonomy of activity recognition on mobile phones. BioNanoScience, 2013,3(2): 145-171

4. Parviainen J, Bojja J, Collin J, et al. Adaptive activity and environment recognition for mobile phones. Sensors, 2014, 14(11): 20753-20778

5. Lee Y S, Cho S B. Activity recognition with android phone using mixture-of-experts co-trained with labeled and unlabeled data. Neurocomputing, 2014, 126: 106-115

6. Deng W Y, Zheng Q H, Wang Z M. Cross-person activity recognition using reduced kernel extreme learning machine. Neural Networks, 2014, 53: 1-7

7. Büber E, Guvensan A M. Discriminative time-domain features for activity recognition on a mobile phone. Proceedings of the IEEE 9th International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP’14), Apr 21-24, 2014, Singapore. Piscataway, NJ, USA: IEEE, 2014: 6p

8. Zeng M, Nguyen L T, Yu B, et al. Convolutional neural networks for human activity recognition using mobile sensors. Proceedings of the IEEE 6th International Conference on Mobile Computing, Applications and Services (MobiCASE’14), Nov 6-7, 2014, Austin, TX, USA. Piscataway, NJ, USA: IEEE, 2014: 197-205

9. Wang Z M, Cao D. Applications of coordinate transformation in mobile user activity recognitions. Journal of Beijing University of Posts and Telecommunications, 2014, 37(4): 30-34 (in Chinese)

10. Zhu Y D, Wang C H, Zhang J Z, et al. Human activity recognition based on similarity. Proceedings of the IEEE 17th International Conference on Computational Science and Engineering (CSE’14), Dec 19-21, 2014, Chengdu, China. Piscataway, NJ, USA: IEEE, 2014: 1382-1387

11. Wang C H, Zhang J Z, Wang Z C, et al. Position-independent activity recognition model for smartphone based on frequency domain algorithm. Proceedings of the IEEE 3rd International Conference on Computer Science and Network Technology (ICCSNT’13), Oct 12-13, 2013, Dalian, China. Piscataway, NJ, USA: IEEE, 2013: 396-399

12. Donoho D L. Compressed sensing. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306

13. Zhang M, Sawchuk A A. Human daily activity recognition with sparse representation using wearable sensors. IEEE Journal of Biomedical and Health Informatics, 2013, 17(3): 553-560

14. Akimura D, Kawahara Y, Asami T. Compressed sensing method for human activity sensing using mobile phone accelerometers. Proceedings of the 9th International Conference on Networked Sensing Systems (INSS’12), Jun 11-14, 2012, Antwerp, Belgium. Piscataway, NJ, USA: IEEE, 2012: 4p

15. Xu W Y, Zhang M, Sawchuk A A, et al. Co-recognition of human activity and sensor placement via compressed sensing in wearable body sensor networks. Proceedings of the 9th International Conference on Wearable and Implantable Body Sensor Networks (BSN’12), May 9-12, 2012, London, UK. Piscataway, NJ, USA: IEEE, 2012: 124-129

16. Song H, Wang Z M. Activity recognition with mobile phone accelerometers by using sparse matrix dictionary method. Application Research of Computers, 2015, 32(9): 2590-2592 (in Chinese)

Candes E, Romberg J. L1-magic: recovery of sparse signals via convex programming. Pasadena, CA, USA: Caltech, 2005

Compressed sensing method for human activity recognition using tri-axis accelerometer on mobile phone

Compressed sensing method for human activity recognition using tri-axis accelerometer on mobile phone

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价