Image Registration using Local Invariant Relational Gabor Features for Automated Underwater Images Mosaicing

Authors: , , ,

Abstract: This paper describes a novel image registration in mosaic framework based on a new type of local invariant feature. Our features are located using Gabor-based features detector which gives spatial distribution of features. We sample a feature descriptor using a newly proposed Relational Gabor Features that adapts relational features onto Gabor magnitude using varying patch sizes obtained during features detection. We tested our proposed features for mosaicing underwater image sequences at ten meters depth taken in Redang Island, Terengganu, Malaysia. Based on the experimental results, the proposed method has produced in highly effective image registration method to obtain a coherent underwater image mosaic.

International Journal of Electrical Engineering and Computer Science, E-ISSN: 2769-2507, Volume 3, 2021, Art. #2

PDF
XML
Cite
×
Citation Tools
WSEAS AMA
Wan Nural Jawahir Hj Wan Yussof, Muhammad Suzuri Hitam, Ezmahamrul Afreen Awalludin, Zainudin Bachok, "Image Registration using Local Invariant Relational Gabor Features for Automated Underwater Images Mosaicing," International Journal of Electrical Engineering and Computer Science, vol. 3, pp. 6-12, 2021, DOI:
Wan Nural Jawahir Hj Wan Yussof, Muhammad Suzuri Hitam, Ezmahamrul Afreen Awalludin, Zainudin Bachok. Image Registration using Local Invariant Relational Gabor Features for Automated Underwater Images Mosaicing. International Journal of Electrical Engineering and Computer Science. 2021;3:6-12.
Copy to Clipboard
Citation copied to Clipboard

Certification

---