An Adaptive Hidden Markov Model for Activity Recognition Based on a Wearable Multi-Sensor Device

详细信息    查看全文

• 作者：Zhen Li (1) (2)
  Zhiqiang Wei (1)
  Yaofeng Yue (2) (3)
  Hao Wang (2) (3)
  Wenyan Jia (2) (3)
  Lora E. Burke (4)
  Thomas Baranowski (5)
  Mingui Sun (2) (2) (3)
  
  1. Department of Computer Science ; Ocean University of China ; Qingdao ; China
  2. Department of Neurosurgery ; University of Pittsburgh ; Pittsburgh ; PA ; USA
  3. Department of Electrical & Computer Engineering ; University of Pittsburgh ; Pittsburgh ; PA ; USA
  4. Department of Health and Community Systems ; University of Pittsburgh ; Pittsburgh ; PA ; USA
  5. Department of Pediatrics ; Baylor College of Medicine ; Houston ; TX ; USA
  
• 关键词：Hidden Markov model ; Activity recognition ; Wearable device ; Big data ; Machine learning ; Personal health
• 刊名：Journal of Medical Systems
• 出版年：2015
• 出版时间：May 2015
• 年：2015
• 卷：39
• 期：5
• 全文大小：3,778 KB
• 参考文献：1. Touati, F, Tabish, R (2013) U-healthcare system: state-of-the-art review and challenges. J. Med. Syst. 37: pp. 1-20 CrossRef
  2. Gyllensten, IC, Bonomi, AG (2011) Identifying types of physical activity with a single accelerometer: evaluating laboratory-trained algorithms in daily life. IEEE Trans. Biomed. Eng. 58: pp. 2656-2663 CrossRef
  3. Khan, AM, Lee, Y-KLY-K, Lee, SY, Kim, T-SKT-S (2010) A triaxial accelerometer-based physical-activity recognition via augmented-signal features and a hierarchical recognizer. IEEE Trans. Inf. Technol. Biomed. 14: pp. 1166-1172 CrossRef
  4. Gao, L., Bourke, A. K., Nelson, J., Activity recognition using dynamic multiple sensor fusion in body sensor networks. In: / 2012 Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. 49:1077鈥?080, 2012.
  5. Chang, S-Y, Lai, C-F, Chao, H-C (2011) An environmental-adaptive fall detection system on mobile device. J. Med. Syst. 35: pp. 1299-1312 CrossRef
  6. Jin, GH, Lee, SB, Lee, TS (2008) Context awareness of human motion states using accelerometer. J. Med. Syst. 32: pp. 93-100 CrossRef
  7. Arab, L, Winter, A (2010) Automated camera-phone experience with the frequency of imaging necessary to capture diet. J. Am. Diet. Assoc. 110: pp. 1238-1241 CrossRef
  8. Kwapisz, JR, Weiss, GM, Moore, SA (2011) Activity recognition using cell phone accelerometers. ACM SIGKDD Explor. Newsl. 12: pp. 74-82 CrossRef
  9. Doherty, A. R., 脫 Conaire, C., Blighe, M., et al., Combining image descriptors to effectively retrieve events from visual lifelogs. In: / Proc. 1st ACM Int. Conf. Multimed. Inf. Retr. 10鈥?7, 2008.
  10. Chatzichristofis, S. A., Boutalis, Y. S., CEDD: color and edge directivity descriptor . A compact descriptor for image indexing and retrieval. In: / Proc. 6th Int. Conf. Comput. Vis. Syst. 312鈥?22, 2008.
  11. Lowe, D. G., Object recognition from local scale-invariant features. In: / Proc. Seventh IEEE Int. Conf. Comput. Vis. 2:1150鈥?157, 1999.
  12. Duan, L, Xu, D, Tsang, IW-H, Luo, J (2012) Visual event recognition in videos by learning from Web data. IEEE Trans. Pattern Anal. Mach. Intell. 34: pp. 1667-1680 CrossRef
  13. Bebars, A. A., Hemayed, E. E., Comparative study for feature detectors in human activity recognition. In: / 2013 9th Int. Comput. Eng. Conf. 19鈥?4, 2013.
  14. Hassan, S. M., Al-Sadek, A. F., Hemayed, E. E., Rule-based approach for enhancing the motion trajectories in human activity recognition. In: / 2010 10th Int. Conf. Intell. Syst. Des. Appl. 829鈥?34, 2010.
  15. Boutell, M, Brown, C (2006) Pictures are not taken in a vacuum - an overview of exploiting context for semantic scene content understanding. IEEE Signal Process. Mag. 23: pp. 101-114
  16. Arif, M, Bilal, M, Kattan, A, Ahamed, SI (2014) Better physical activity classification using smartphone acceleration sensor. J. Med. Syst. 38: pp. 1-10 CrossRef
  17. Yin, J, Yang, Q, Pan, JJ (2008) Sensor-based abnormal human-activity detection. IEEE Trans. Knowl. Data Eng. 20: pp. 1082-1090 CrossRef
  18. Giansanti, D, Macellari, V, Maccioni, G (2008) New neural network classifier of fall-risk based on the Mahalanobis distance and kinematic parameters assessed by a wearable device. Physiol. Meas. 29: pp. N11-N19 CrossRef
  19. Cao, LCL, Luo, JLJ, Kautz, H, Huang, TS (2009) Image annotation within the context of personal photo collections using hierarchical event and scene models. IEEE Trans. Multimed. 11: pp. 208-219 CrossRef
  20. Hughes, G (1968) On the mean accuracy of statistical pattern recognizers. IEEE Trans. Inf. Theory 14: pp. 55-63 CrossRef
  21. Zhang, W., Jia, W., Sun, M., Segmentation for efficient browsing of chronical video recorded by a wearable device. In: / Proc 2010 I.E. 36th Annu. Northeast. Bioeng. Conf. 1鈥?, 2010.
  22. Manjunath, BS, Ohm, J-R, Vasudevan, VV, Yamada, A (2001) Color and texture descriptors. IEEE Trans. Circ. Syst. Video Technol. 11: pp. 703-715 CrossRef
  23. Csurka, G., Dance, C., Fan, L., et al., Visual categorization with bags of keypoints. In: / Work. Stat. Learn. Comput. Vision ECCV. 1鈥?, 2004.
  24. Peng, H, Long, F, Ding, C (2005) Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans. Pattern Anal. Mach. Intell. 27: pp. 1226-1238 CrossRef
  25. Powers, DM (2011) Evaluation: from precision, recall and F-measure to ROC, informedness, markedness & correlation. J. Mach. Learn. Technol. 2: pp. 37-63
  
• 刊物类别：Mathematics and Statistics
• 刊物主题：Statistics
  Statistics for Life Sciences, Medicine and Health Sciences
  Health Informatics and Administration
  
• 出版者：Springer Netherlands
• ISSN：1573-689X

文摘

Human activity recognition is important in the study of personal health, wellness and lifestyle. In order to acquire human activity information from the personal space, many wearable multi-sensor devices have been developed. In this paper, a novel technique for automatic activity recognition based on multi-sensor data is presented. In order to utilize these data efficiently and overcome the big data problem, an offline adaptive-Hidden Markov Model (HMM) is proposed. A sensor selection scheme is implemented based on an improved Viterbi algorithm. A new method is proposed that incorporates personal experience into the HMM model as a priori information. Experiments are conducted using a personal wearable computer eButton consisting of multiple sensors. Our comparative study with the standard HMM and other alternative methods in processing the eButton data have shown that our method is more robust and efficient, providing a useful tool to evaluate human activity and lifestyle.