Design and implementation of a convolutional neural network on an edge computing smartphone for human activity recognition
- Submitting institution
-
The University of East Anglia
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 182622254
- Type
- D - Journal article
- DOI
-
10.1109/ACCESS.2019.2941836
- Title of journal
- IEEE Access
- Article number
- -
- First page
- 133509
- Volume
- 7
- Issue
- -
- ISSN
- 2169-3536
- Open access status
- Compliant
- Month of publication
- October
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
4
- Research group(s)
-
-
- Citation count
- 8
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper served as a deployable solution for the implementation of the ‘behavioural phenotyping’ part of EPSRC funded Wearable Clinic: Connecting Health, Self, and Care (EP/P010148/1) project. An android app was built using the findings from this research, contributing to an Open Source Toolset for doing AI on edge devices. The developed application recognizes any concerning user physical activity using an on-device deep learning model and communicates the results for follow-up intervention in real-time. This context-aware prototype system works as an integrated proof of concept for secure collection of data, performs low-power model-driven inference to maintain user privacy.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -