Abstract

Fine art is usually produced on paper or canvas as a one-of-a-kind artwork. Fine art may be reproduced in limited editions and put up for sale as art. Different printing technologies have been used in fine art reproduction such as lithography, screenprinting, and most recently inkjet.

The research aspect of watercolor reproduction has been the question of "how good is good enough." In this case, the artists demand the exact match between the original watercolor and its reproduction. While there are difficulties in quantifying the degree of color image match, the initial testing of watercolor reproduction using a colormanaged approach with an inkjet printer showed that there is a need to improve the reproduction quality. The objective of this study is to see if accuracy of watercolor reproduction can be improved by using profile editing tools.

The significance of the research is the potential to achieve higher reproduction quality in watercolor by means of profile editing. In addition, we can put control back in the hands of content creators for limited editions.

This research begins with a literature review. The review discusses how artworks are being digitized and reproduced by museums. It points out the wide adoption of International Color Consortium (ICC) color management practices in printing and publishing. It also covers how a color image match between an original and its reproduction is assessed quantitatively and qualitatively.

The quantitative analyses of Macbeth ColorChecker between a generic ICC profile and a custom ICC profile were used to test first hypothesis, i.e., if there is any significant difference in measured color accuracy of watercolor reproductions between a generic ICC profile and a custom ICC profile. The results indicate that there is a significant difference in color accuracy of watercolor reproduction between using generic ICC profile and the custom ICC profile state findings. To our surprise, the custom ICC profile performed worse than the generic ICC profile. A possible cause of the large color differences was attributed to the accuracy of the scanner profile.

A paired comparison method was used to test the second hypothesis, i.e., if there is any significant visual difference in color accuracy of watercolor reproductions between an unedited ICC profile and an edited ICC profile. The results indicate that there is no significant difference in color accuracy of watercolor reproduction between an unedited ICC profile and an edited ICC profile state findings. To our surprise once again, edited profiles did not perform color matching any better than unedited profiles. A major factor is that editing of tone reproduction and gray balance are treated as two separate events in the profile editing process. In fact, tone reproduction and gray balance are dependent on each other.

Library of Congress Subject Headings

Image processing--Digital techniques; Art objects--Reproduction; Color printing--Data processing; Watercolor painting

Publication Date

2006

Document Type

Thesis

Student Type

Graduate

Degree Name

Print Media (MS)

Department, Program, or Center

School of Print Media (CIAS)

Advisor

Robert Chung

Advisor/Committee Member

Frank Cost

Comments

Physical copy available from RIT's Wallace Library at TA1637 .K56 2006

Campus

RIT – Main Campus

Download

Share

COinS