Abstract

Original and reproduced art are usually viewed under quite different viewing conditions. One of the interesting differences in viewing condition is size difference. The main focus of this research was investigation of the effect of image size on color perception of rendered images. This research had several goals. The first goal was to develop an experimental paradigm for measuring the effect of image size on color appearance. The second goal was to identify the most affected image attributes for changes of image size. The final goal was to design and evaluate algorithms to compensate for the change of visual angle (size). To achieve the first goal, an exploratory experiment was performed using a colorimetrically characterized digital projector and LCD. The projector and LCD were light emitting devices and in this sense were similar soft-copy media. The physical sizes of the reproduced images on the LCD and projector screen could be very different. Additionally, one could benefit from flexibility of soft-copy reproduction devices such as real-time image rendering, which is essential for adjustment experiments. The capability of the experimental paradigm in revealing the change of appearance for a change of visual angle (size) was demonstrated by conducting a paired-comparison experiment. Through contrast matching experiments, achromatic and chromatic contrast and mean luminance of an image were identified as the most affected attributes for changes of image size. Measurement of the extent and trend of changes for each attribute were measured using matching experiments. Proper algorithms to compensate for the image size effect were design and evaluated. The correction algorithms were tested versus traditional colorimetric image rendering using a paired-comparison technique. The paired-comparison results confirmed superiority of the algorithms over the traditional colorimetric image rendering for the size effect compensation.

Library of Congress Subject Headings

Imaging systems--Image quality--Data processing; Image processing--Digital techniques; Art--Reproduction; Colorimetric analysis; Color vision

Publication Date

5-1-2008

Document Type

Dissertation

Student Type

Graduate

Department, Program, or Center

Chester F. Carlson Center for Imaging Science (COS)

Advisor

Fairchild, Mark

Advisor/Committee Member

Johnson, Garrett

Advisor/Committee Member

Lent, Tina

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: TA1637 .N49 2008

Campus

RIT – Main Campus

Download

Share

COinS