A Compositional Language for Color Based Image Retrieval

Many systems and techniques have been proposed so far to support image retrieval based on color content. However, in all these experiences only low-level color properties are considered. High-level color properties address concepts such as the perceptual quality of colors and the sensations that they convey. Color-induced sensations include warmth, accordance or contrast, harmony, excitement, depression, anguish etc. Paintings are an example where the visual content is related more to high-level color qualities and spatial arrangements than to physical properties of colors. Starting from this observation, Johannes Itten introduced a formalism to analyze the use of color in art and the effects that this induces on the user's psyche. In this paper, we present a system which translates the Itten theory into a formal language expressing the semantics associated with the combination of chromatic properties of color regions. A formal language and a model checking engine are implemented to define semantic clauses and verify the degree of truth by which they hold over an image.