Author

Ellen A. Day

Abstract

Two paired-comparison psychophysical experiments were performed. The stimuli consisted of six image types resultingfrom several multispectral image-capture and reconstruction techniques. A seventh image type, color-managed images from a high-end consumer camera, was also included in thefirst experiment to compare the accuracy of commercial RGB imaging. The images were evaluated under simulated daylight (6800K) and incandescent (2700K) illumination. The first experiment evaluated color reproduction accuracy. Under simulated daylight, the subjects judged all of the images to have the same color accuracy, except the consumer camera image which was significantly worse. Under incandescent illumination, all the images, including the consumer camera, had equivalent performance. The second experiment evaluated image quality. The results of this experiment were highly target dependent. A subsequent image registration experiment showed that the results of the image quality experiment were affected by image registration to some degree. An analysis of the color reproduction accuracy and image quality experiments combined showed that the consumer camera image type was preferred the least over all. The most preferred image types were the thirty-one-channel image type and both six-channel image types created using RGB filters along with a Wratten filter, with eigenvector analysis and pseudo-inverse transformations.

Library of Congress Subject Headings

Photography--Digital techniques; Image processing--Digital techniques; Imaging systems--Image quality; Color sensitometry (Photogrpahy); Images, Photographic; Art--Digitization

Publication Date

2003

Document Type

Thesis

Student Type

Graduate

Degree Name

Color Science (MS)

Department, Program, or Center

Chester F. Carlson Center for Imaging Science (COS)

Advisor

Roy Berns

Advisor/Committee Member

Ethan Montag

Comments

Physical copy available from RIT's Wallace Library at TR267 .D39 2003

Campus

RIT – Main Campus

Share

COinS