|
HUPERSPECTRAL DATA ANALYSIS
Spectrum shape elements model for correction of multichannel images
A. V. Nikonorovab a Samara State Aerospace University
b Image Processing Systems Institute, Russian Academy of Sciences
Abstract:
This paper presents the spectral-shape elements model for correction of non-isoplanatic deviation in the scene illumination. I propose an identification of the correction function on the set of spectral shape elements with the Hausdorff metric. Also a necessary condition which allows to obtain an adequate form of the color correction function is presented. The experiments performed on real images confirm the high quality of the proposed color correction technique with respect to well-known Retinex method.
Keywords:
image processing, color correction, dichromatic model, spectral shape elements, Hausdorff distance, non-negative LSM, Retinex model, color constancy.
Received: 12.02.2014
Citation:
A. V. Nikonorov, “Spectrum shape elements model for correction of multichannel images”, Computer Optics, 38:2 (2014), 304–313
Linking options:
https://www.mathnet.ru/eng/co276 https://www.mathnet.ru/eng/co/v38/i2/p304
|
| Statistics & downloads: |
| Abstract page: | 209 | | Full-text PDF : | 94 | | References: | 53 |
|