Share Email Print
cover

Proceedings Paper

LabRGB: optimization of bit allocation
Author(s): Fumio Nakaya
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Spectral distribution can be written as a linear combination of eigenvectors and the eigenvectors method gives the least estimation error, but eigenvectors depend on a sample selection of population and encoding values have no physical meaning. Recently reported LabPQR [1] is to convey physical values, but still is dependent on a sample selection of population. Thus, LabRGB [2][3][4] was proposed in 2007. LabRGB is to provide "sample selection of population" free spectral encoding/decoding methods. LabRGB consists of six unique trigonometric base functions and physically meaningful encoding values. LabRGB was applied to the real multispectral images and showed almost equal performance to traditional orthogonal eigenvector method in spectral estimation, and even better performance in colorimetric estimation. In this paper, bit allocation to the weighting factors were examined in terms of spectral and colorimetric distance of nearest neighbors. The optimum way of minimizing the unusable combination of weighting factors were obtained by using the correlation of the weighting factors. The optimum way of minimizing the spectral and colorimetric distance of nearest neighbors was also obtained by using the nonlinear mapping method. The two methods thus obtained give a good clue for explicitly defining the number of bits of respective scores for the future applications and standardization.

Paper Details

Date Published: 25 January 2011
PDF: 7 pages
Proc. SPIE 7866, Color Imaging XVI: Displaying, Processing, Hardcopy, and Applications, 78660Y (25 January 2011); doi: 10.1117/12.871716
Show Author Affiliations
Fumio Nakaya, Fuji Xerox Co., Ltd. (Japan)


Published in SPIE Proceedings Vol. 7866:
Color Imaging XVI: Displaying, Processing, Hardcopy, and Applications
Reiner Eschbach; Gabriel G. Marcu; Alessandro Rizzi, Editor(s)

© SPIE. Terms of Use
Back to Top