Frequency domain subpixel registration using HOG phase correlation
- Submitting institution
-
Kingston University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11-04-1338
- Type
- D - Journal article
- DOI
-
10.1016/j.cviu.2016.10.019
- Title of journal
- Computer Vision and Image Understanding
- Article number
- -
- First page
- 70
- Volume
- 155
- Issue
- -
- ISSN
- 1077-3142
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2017
- 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
- 5
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work introduces a method for frequency-domain image registration based on image filtering using dense Histogram of Oriented Gradients. The proposed approach offers high registration accuracy for global and local motions even with only small overlapped regions. It outperforms the state-of-the-art in frequency-domain motion estimation and supports large image sizes with large translations and rotations. As a results it is suitable, not only for medical images such as MRI registration, but also for satellite images offering more accurate alignment helping to create maps automatically and detect changes such as building changes in the environment.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -