8 December 1995 Image set compression with eigenvectors spectral representation
Author Affiliations +
Proceedings Volume 2605, Coding and Signal Processing for Information Storage; (1995) https://doi.org/10.1117/12.228229
Event: Photonics East '95, 1995, Philadelphia, PA, United States
In this work the new spectral representation for noisy images is investigated. The orthogonal image expansion is based on the eigenvectors of symmetric CK-criterion which takes into account both correlation properties of signal and noise ensembles. We suppose that such representation provides maximal concentration of noiseless image in main spectral components, and so the effective noise suppression under spectrum truncation can be achieved. An effective algorithm for KL and CK-eigenvectors calculation has been proposed. Performed numerical experiments verify the algorithm efficiency and its suitability for image processing problems.
© (1995) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Alexandr G. Chumakov, Alexandr G. Chumakov, Vitalij N. Kurashov, Vitalij N. Kurashov, } "Image set compression with eigenvectors spectral representation", Proc. SPIE 2605, Coding and Signal Processing for Information Storage, (8 December 1995); doi: 10.1117/12.228229; https://doi.org/10.1117/12.228229

Back to Top