Back to articles
Articles
Volume: 1 | Article ID: art00028
Image
Color Rendition with Variable Point Interpolation
Henry Chen
King Lung Huang
  DOI :  10.2352/CIC.1993.1.1.art00028  Published Online :  January 1993
Abstract

Because of the rapid growth in electronic color publishing, it becomes more important to develop a color correct algorithm to minimize the color errors caused from the gamut mismatch, quantization and non-linearity characteristics between image devices. Recently, some 3-D (3-Dimensional) interpolation algorithms were introduced and most of them are lacking the flexibility of adopting different LUTs (Look Up Table) and interpolation algorithms to correct the non-uniformity of color space and to fit the reception of human vision. This paper presents a new algorithm of color transformation with non-fixed points interpolation method to deal with color correction in non-uniform color space.In this algorithm, the non-uniformity of color space is associated with the membership function of modern fuzzy theory. Based on the association of fuzzy rules, color rendition can be achieved with LUTs and mixed mode variable point interpolation method. According to the pictorial image coordinate in the non-uniform color space, different LUTs and 1, 2, 3, 4, 5, 6, 7, or 8 points interpolation will be selected. By considering the nonlinearity and the different resolution requirement of color correction under the human appearance-based model, the color reproduction model is established in the LUT. With the fulfillment of the scheme, it can reduce the computational costs by a large factor comparing the conventional direct color mapping method and it can improve the color matching accuracy of the 3-D interpolation techniques. The colorimetric simulation of this algorithm illustrates a satisfac-tory performance. It is estimated that mean square color difference, delta Erms, between two color images will be less than 6. This algorithm is verified by various output samples produced by this new method. The color printing samples using this new algorithm are compared with the outputs using traditional 3-D interpolation technique and the results indicates a great improvement. It is proved to be a less cost and more flexible scheme for electronic color publishing.

Journal Title : Color and Imaging Conference
Publisher Name : Society of Imaging Science and Technology
Publisher Location : 7003 Kilworth Lane, Springfield, VA 22151, USA
Subject Areas :
Views 4
Downloads 0
 DOI 10.2352/CIC.1993.1.1.art00028
 articleview.views 4
 articleview.downloads 0
  Cite this article 

Henry Chen, King Lung Huang, "Color Rendition with Variable Point Interpolation"  in Proc. IS&T 1st Color and Imaging Conf.,  1993,  pp 113 - 117,  https://doi.org/10.2352/CIC.1993.1.1.art00028

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 1993
72010350
Color and Imaging Conference
color imaging conf
2166-9635
Society of Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151, USA
2166-9635(19930101)1993:1L.113;1-
cic_v1993n1/splitsection28.xml
/ist/cic/1993/00001993/00000001/art00028
Articles
Color Rendition with Variable Point Interpolation
ChenHenry
HuangKing Lung
01011993
1993
1
113
117
1993
Because of the rapid growth in electronic color publishing, it becomes more important to develop a color correct algorithm to minimize the color errors caused from the gamut mismatch, quantization and non-linearity characteristics between image devices. Recently, some 3-D (3-Dimensional) interpolation algorithms were introduced and most of them are lacking the flexibility of adopting different LUTs (Look Up Table) and interpolation algorithms to correct the non-uniformity of color space and to fit the reception of human vision. This paper presents a new algorithm of color transformation with non-fixed points interpolation method to deal with color correction in non-uniform color space.In this algorithm, the non-uniformity of color space is associated with the membership function of modern fuzzy theory. Based on the association of fuzzy rules, color rendition can be achieved with LUTs and mixed mode variable point interpolation method. According to the pictorial image coordinate in the non-uniform color space, different LUTs and 1, 2, 3, 4, 5, 6, 7, or 8 points interpolation will be selected. By considering the nonlinearity and the different resolution requirement of color correction under the human appearance-based model, the color reproduction model is established in the LUT. With the fulfillment of the scheme, it can reduce the computational costs by a large factor comparing the conventional direct color mapping method and it can improve the color matching accuracy of the 3-D interpolation techniques. The colorimetric simulation of this algorithm illustrates a satisfac-tory performance. It is estimated that mean square color difference, delta Erms, between two color images will be less than 6. This algorithm is verified by various output samples produced by this new method. The color printing samples using this new algorithm are compared with the outputs using traditional 3-D interpolation technique and the results indicates a great improvement. It is proved to be a less cost and more flexible scheme for electronic color publishing.