Abstract

In this thesis, a paradigm was generated for color gamut mapping of pictorial images. This involved the development and testing of: 1.) a hue-corrected version of the CIELAB color space, 2.) an image-dependent sigmoidal-lightness-rescaling process, 3.) an image-gamut- based chromatic-compression process, and 4.) a gamut-expansion process. This gamut-mapping paradigm was tested against some gamut-mapping strategies published in the literature. Reproductions generated by gamut mapping in a hue-corrected CIELAB color space more accurately preserved the perceived hue of the original scenes compared to reproductions generated using the CIELAB color space. The results of three gamut-mapping experiments showed that the contrast-preserving nature of the sigmoidal-lightness-remapping strategy generated gamut-mapped reproductions that were better matches to the originals than reproductions generated using linear-lightness-compression functions. In addition, chromatic-scaling functions that compressed colors at a higher rate near the gamut surface and less near the achromatic axis produced better matches to the originals than algorithms that performed linear chroma compression throughout color space. A constrained gamut-expansion process, similar to the inverse of the best gamut-compression process found in this experiment, produced reproductions preferred over an expansion process utilizing unconstrained linear expansion.

Library of Congress Subject Headings

Color computer graphics; Color printing--Digital techniques; Image processing--Digital techniques

Publication Date

10-1-1999

Document Type

Dissertation

Student Type

Graduate

Department, Program, or Center

Chester F. Carlson Center for Imaging Science (COS)

Advisor

Fairchild, Mark

Advisor/Committee Member

Berns, Roy

Advisor/Committee Member

Balasubramanian, Raja

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works. Physical copy available through RIT's The Wallace Library at: T385 .B72 1999

Campus

RIT – Main Campus

Share

COinS