A prior regularized multi-layer graph ranking model for image saliency computation
- Submitting institution
-
University of Greenwich
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 24364
- Type
- D - Journal article
- DOI
-
10.1016/j.neucom.2018.06.072
- Title of journal
- Neurocomputing
- Article number
- -
- First page
- 234
- Volume
- 315
- Issue
- -
- ISSN
- 0925-2312
- Open access status
- Deposit exception
- Month of publication
- -
- Year of publication
- 2018
- 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
-
4
- Research group(s)
-
-
- Citation count
- 3
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The prior regularised multi-layer graph ranking model proposed uses the prior calculation by boundary connectivity, and overcomes the failure of previous MR models in considering the prior information in its ranking process. Conducted on four public databases (1000-image ECSSD, 1000-image ASD, 200-image SED, and 850-image PASCAL-S) with regards to Precision-Recall curves, F-measure and Mean Absolute Error, the experimental results show that the novel model outperforms those state-of-the-art methods (15 were compared) at the time the paper was produced. This work is sponsored by the National Natural Science Foundation of China, Anhui Higher Education Institution and Anhui University (China).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
