Abstract :
To identify the real time activities, an online algorithm need be considered. In this paper, we will first segment entire one activity as one time interval using Bayesian online detection method instead of fixed and small length time interval. Then, we introduce two layer random forest classification for real time activity recognition on the smartphone by embedded accelerometers. We evaluate the performance of our method based on six activities walking, upstairs, downstairs, sitting, standing, and laying on 30 volunteers. For the data considered, we get 92.4 overall accuracy based on six activities and 100 overall accuracy only based on dynamic activity and static activity. Shuang Na | Kandethody M. Ramachandran | Ming Ji | Yicheng Tu "Real-time Activity Recognition using Smartphone Accelerometer" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-1 , December 2019, URL: https://www.ijtsrd.com/papers/ijtsrd29550.pdf Paper URL: https://www.ijtsrd.com/mathemetics/other/29550/real-time-activity-recognition-using-smartphone-accelerometer/shuang-na
Keyword :
Bayesian online detection, Human activity recognition, Random forest