Car make and model recognition under limited lighting conditions at night
- Submitting institution
-
Bournemouth University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 211486
- Type
- D - Journal article
- DOI
-
10.1007/s10044-016-0559-6
- Title of journal
- Pattern Analysis and Applications
- Article number
- 0
- First page
- 1195
- Volume
- 20
- Issue
- 4
- ISSN
- 1433-7541
- Open access status
- Compliant
- Month of publication
- May
- Year of publication
- 2016
- 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
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- 9
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Most car make and model recognition systems perform well in well-lit environments. There are not many publications that address the matter in minimal lighting conditions. Most features of a vehicle disappear, only some vehicle’s lights can be seen. The paper addresses the issue using an ensemble of classical machine learning techniques, with limited data, producing highly accurate results without using data-hungry deep learning. Since the paper has addressed the specific need very well, it has led to the work with a charity campaigning for justice to analyse video evidence, in a criminal case, taken under very limited lighting condition.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -