A local segmentation structure is proposed for color image denoising and restoration. Most state-of-the-art filters suppress impulse noise well but tend to destroy thin lines, edges, and high image details. The proposed filters facilitate local segmentation technique to preserve image structures and noise reduction. Firstly, the K-VMF is used for local segmentation and then the selection of vector filters for reconstruction of the current pixel. The proposed filters demonstrated acceptable results for both objective and subjective assessments.
History
Start page
589
End page
595
Total pages
7
Outlet
Proceedings of the Ninth International Conference on Knowledge-Based Intelligent Information and Engineering Systems
Editors
J. Carbonell, J. Siekmann
Name of conference
International Conference on Knowledge-Based Intelligent Information and Engineering Systems