Applications of vector space methods to the colorimetric problems of scanning filter selection and gamut mismatch compensation
Abstract
This thesis consists of 3 independent chapters. Chapter 1 is theoretical. It introduces the reader to colorimetry through the use of vector space concepts. Its first use is as a tutorial for the engineer/mathematician who is interested in basic principles of color science. Its alternate use is to introduce modern methods of analysis to those persons familiar with colorimetry. Chapter 2 considers the problem of designing scanning filters for a color measurement device. The device is meant to characterize both luminous and reflective surfaces under either of 2 standard illuminants. It uses an internal set of light emitting diodes as illumination sources but gives accurate surface characterizations for both EE and D65 conditions. An optimization method is presented which minimizes the perception of color shifts due to non-ideal filters, while maintaining filter smoothness and robustness to noise. Chapter 3 addresses the gamut mapping problem. The gamut of a device is a volume representing the colors which the device is capable of producing. We present a framework for specifying image and device gamuts, and develop a methodology for lightness-saturation mapping which tends to retain lightness and contrast of achromatic pixels while preserving hue and saturation of chromatic pixels. The mapping results in a marked improvement of display image quality over methods commonly used for displaying photo-CD images.
Degree
Ph.D.
Advisors
Allebach, Purdue University.
Subject Area
Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.