Abstract
With the recent spread of mobile devices equipped with different sensors, it is possible to continuously recognise and monitor activities in daily life. This sensor-based human activity recognition is formulated as sequence classification to categorise sequences of sensor values into appropriate activity classes. One crucial problem is how to model features that can precisely represent characteristics of each sequence and lead to accurate recognition. It is laborious and/or difficult to hand-craft such features based on prior knowledge and manual investigation about sensor data. To overcome this, we focus on a feature learning approach that extracts useful features from a large amount of data. In particular, we adopt a simple but effective one, called codebook approach, which groups numerous subsequences collected from sequences into clusters. Each cluster centre is called a codeword and represents a statistically distinctive subsequence. Then, a sequence is encoded as a feature expressing the distribution of codewords. The extensive experiments on different recognition tasks for physical, mental and eye-based activities validate the effectiveness, generality and usability of the codebook approach.
Highlights
Mobile devices equipped with different sensors have been made low-power, low-cost, high-capacity and miniaturised [1,2,3]
Descriptions that can be potentially extracted from ”low-level” sensor data. This sensor-based human activity recognition is useful for various applications, like intelligent human-computer interaction, effective lifelogging and healthcare
Except RESpiration spectra (RES_sp), the remaining five plots indicates the accuracy distributions resulting from soft assignment with σ = 0.125, 0.25, 0.5, 1, 2
Summary
Mobile devices equipped with different sensors have been made low-power, low-cost, high-capacity and miniaturised [1,2,3] This allows continuous recording of sensor data related to the daily life of a person [4,5]. In addition to traditional accelerometers, gyroscopes and GPSs, state-of-the-art sensors can capture physiological signals such as blood volume pressure, heart rate, galvanic skin conductance, respiration rate and electrooculogram (EOG) [6,7,8,9,10] This offers a possibility to recognise physical behaviours of the person, and his/her mental states (i.e., emotions) and health conditions. This sensor-based human activity recognition is useful for various applications, like intelligent human-computer interaction, effective lifelogging and healthcare (ambient-assisted living)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
