Current color image encoding and processing generally considers the color value of a pixel based on a quantized set of coordinates in an n-dimensional color space. While this approach is useful and practical for numerous reasons, it is informative to consider how these representations and processing techniques might make use of the highest level cognitive description of color, the color name or color category. This paper proposes how a lexical or name-based representation of an image might be generated. Color naming algorithms based on classical logic, prototype theory and fuzzy logic are presented and compared. This paper also describes specific applications of this lexical quantization such as morpholexical analysis, lexical harmony and non-photorealistic rendering.