Scene perception guided crowd anomaly detection
- Submitting institution
-
University of Portsmouth
- Unit of assessment
- 12 - Engineering
- Output identifier
- 22015424
- Type
- D - Journal article
- DOI
-
10.1016/j.neucom.2020.07.019
- Title of journal
- Neurocomputing
- Article number
- 0
- First page
- 291
- Volume
- 414
- Issue
- -
- ISSN
- 0925-2312
- Open access status
- Compliant
- Month of publication
- July
- Year of publication
- 2020
- 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
-
5
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This interdisciplinary work demonstrates a novel synergy of graphical features and fluidic dynamics features using existing imagery data set free to the public. The paper is significant because it enhances accuracy and optimal computational cost providing a distinct advantage over existing algorithms for dynamic crowd anomaly surveillance. The devised framework can be applied to any dynamic scenes combining spatial and temporal features, and to extract subject attributes as demonstrated by further studies by Hoang et al. (10.1109/ACCESS.2021.3053072) for crowd counting, Zhang et al.(10.3390/rs12203316) for ship detection.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -