Color image processing for high-quality reproduction based on uniform color spaces

Color can increase the information provided by an image in addition to enhancing its aesthetic value. As a result, many products and services have evolved from a monochrome to a color environment during the past several decades. Examples include photography and motion pictures, television and video display devices, copying systems, facsimile, and printed products such as newspapers, magazines, and brochures. The areas of quantization, halftoning, gamut mismatch compensation, and data compression have been investigated extensively for monochrome imagery. The ongoing evolution from monochrome to color of many products and services has focused new interest on these research areas as applied to a multidimensional or color environment. Often these algorithms consider the color image as a separable collection of monochrome images processing each in an independent and identical manner. This approach disregards the color sensitivity of the human viewer as well as the characteristics of the color gamut for real images. In this thesis, we investigate the areas of quantization, halftoning, gamut mismatch compensation, and data compression for color imagery processed to near original image quality. Our techniques incorporate a colorimetric approach and uniform color spaces to account for the human viewers color sensitivity