A Study on Meal Time Estimation and Eating Behavior Recognition Considering Movement Using Wrist-Worn Accelerometer with Its Frequency


KIPS Transactions on Software and Data Engineering, Vol. 6, No. 1, pp. 29-36, Jan. 2017
10.3745/KTSDE.2017.6.1.29,   PDF Download:
Keywords: Eating, Accelerometer, Activity Recognition
Abstract

In this paper, we propose a method for recognizing eating behavior with almost no motion acceleration. First, by using the acceleration of gravity acting on the wrist direction, we calculate the angle between the gravity and the wrist direction. After that, detect wrist reciprocating motion when peak and vally exist in specific angle band. And then, when accumulate the number of wrist reciprocating motion occurrences are up to 10, then regard as the meal time 5 minutes before the detection time. Also, estimate the meal time only if its duration is more than 7 minutes. Using the data of 2128 minutes, which was collected from four graduate student, the result of the meal time estimation shows 95.63% accuracy.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
K. C. Park, S. Choe, W. Cho, "A Study on Meal Time Estimation and Eating Behavior Recognition Considering Movement Using Wrist-Worn Accelerometer with Its Frequency," KIPS Transactions on Software and Data Engineering, vol. 6, no. 1, pp. 29-36, 2017. DOI: 10.3745/KTSDE.2017.6.1.29.

[ACM Style]
Kyeong Chan Park, Sun-Taag Choe, and We-duke Cho. 2017. A Study on Meal Time Estimation and Eating Behavior Recognition Considering Movement Using Wrist-Worn Accelerometer with Its Frequency. KIPS Transactions on Software and Data Engineering, 6, 1, (2017), 29-36. DOI: 10.3745/KTSDE.2017.6.1.29.