Abstract
As an active research field, sport-related activity monitoring plays an important role in people’s lives and health. This is often viewed as a human activity recognition task in which a fixed-length sliding window is used to segment long-term activity signals. However, activities with complex motion states and non-periodicity can be better monitored if the monitoring algorithm is able to accurately detect the duration of meaningful motion states. However, this ability is lacking in the sliding window approach. In this study, we focused on two types of activities for sport-related activity monitoring, which we regard as a human activity detection and recognition task. For non-periodic activities, we propose an interval-based detection and recognition method. The proposed approach can accurately determine the duration of each target motion state by generating candidate intervals. For weak periodic activities, we propose a classification-based periodic matching method that uses periodic matching to segment the motion sate. Experimental results show that the proposed methods performed better than the sliding window method.
Highlights
With the widespread use of wearable smart devices, sport-related activity monitoring using inertial sensors has become an active area of research, and it is used to improve the quality of life and promote personal health
We evaluated the performance of interval generation by verifying the quality of the generated candidate intervals, which were taken as the output
We focus on the detection and recognition of two types of activities, namely, the non-periodic activity with complex motion states (NP_CMS) and the weakly periodic activity with complex motion states (WP_CMS)
Summary
With the widespread use of wearable smart devices, sport-related activity monitoring using inertial sensors has become an active area of research, and it is used to improve the quality of life and promote personal health. The activity monitoring problem has often been defined as a human activity recognition (HAR) task [1]. An HAR model usually consists of two main parts—data segmentation and activity classification. The segmentation procedure uses a fixed-length sliding window to divide the sensor signal into different segments. The subsequent classifier is used to classify these activities by using the information in these segments. Many researchers [2,3,4,5]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call