David Miller


A relationship between preferred tone reproduction and final picture size is suggested in many forms in the literature and in practical experience. This project uses photographic halftone prints varying in tone reproduction and produced at different sizes to investigate this relationship. The tone reproduction variations are introduced electronically using an RCA color scanner and a method of eigenvector analysis is used to reduce these tone reproductions down to their important components. The final set of prints consisting of ten tone reproductions, three reproduced sizes, and two sharpnesses are evaluated to determine their subjective quality in the judgement of a group of ten observers. These quality ratings are normalized and averaged and then mathematically correlated with an eigenvector representation of their tone reproduction. The preferred tone reproduction (producing maximum quality) may then be determined for each picture size. Due to large differences in quality rating from observer to observer and an incomplete selection of tone reproductions, no definite relationship was determined. Useful information was obtained on the use of the eigenvector method, the magnitude of quality variation as tone reproduction is changed, the difficulties of subjective evaluation, and a calculated tone reproduction producing maximum quality for a normal print not considering size.

Library of Congress Subject Headings

Images, Photographic

Publication Date


Document Type


Department, Program, or Center

School of Photographic Arts and Sciences (CIAS)


Yule, J.


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: TR222.M54


RIT – Main Campus