In computer vision, color has always played an important role. Color can provide critical information for a variety of computer vision tasks such as image matching, object recognition, and image retrieval. However, object colors are sensitive to lighting geometry, illumination color, and hightlights. In this paper we present a color model for image matching and object recognition that is invariant for illumination and highlights. The color model is defined as the ratios of the color differences between neighboring pixels for each color component. Based on the dichromatic reflection color model, it is shown that the proposed color model is invariant to lighting geometry, illumination color, highlights, and diffuse lighting. Experimental results show robust image matching using the proposed color model on objects that are illuminated under different illumination colors and lighting geometry.
