Skip to main content
ARS Home » Plains Area » Houston, Texas » Children's Nutrition Research Center » Research » Publications at this Location » Publication #315801
Title: An adaptive Hidden Markov Model for activity recognition based on a wearable multi-sensor device

Author
item LI, ZHEN - Ocean University Of China
item WEI, ZHIQIANG - Ocean University Of China
item YUE, YAOFENG - University Of Pittsburgh
item WANG, HAO - University Of Pittsburgh
item JIA, WENYAN - University Of Pittsburgh
item BURKE, LORA - University Of Pittsburgh
item BARANOWSKI, THOMAS - Children'S Nutrition Research Center (CNRC)
item SUN, MINGUI - University Of Pittsburgh

Submitted to: Journal of Medical Systems
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/2/2015
Publication Date: 5/1/2015
Citation: Li, Z., Wei, Z., Yue, Y., Wang, H., Jia, W., Burke, L.E., Baranowski, T., Sun, M. 2015. An adaptive Hidden Markov Model for activity recognition based on a wearable multi-sensor device. Journal of Medical Systems. 39(5):57.

Interpretive Summary: Since there is much error in self reports of level or type of physical activity, and accelerometers or pedometers provide objective information on intensity, but not type of activity, new methods are needed to automate the assessment of identifying activity. Taking all day images of everything in front of a person is a method being explored by several research groups. The eButton includes the all day camera and an accelerometer. A new procedure was tested using the images and accelerometer combined. The new procedure worked better than the other current systems for automated recognition of physical activity. If this early work is verified in future research, the accuracy of the assessment of physical activity could be substantially enhanced, perhaps revolutionizing our understanding of the role of physical activity in health.

Technical Abstract: 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.