What is a color appearance model?

A color appearance model tries to model how the human visual system perceives the color of an object under different lighting conditions and with different backgrounds. By developing such a model, an image that is viewed under one lighting condition can be adjusted to appear to have the same colors if it were viewed under a completely different lighting condition. The ability of a color appearance model to perform this task means that it can be used to develop a device-independent way of storing images. For example, if an image is scanned into a computer, the color appearance model, knowing the scanner's illumination conditions, can transform the image into its internal representation. When the image would be displayed on a monitor or printed, again the color appearance, knowing the output display's illumination conditions, can transform its internal representation on the image to the correct representation for the output display.

What is CIECAM97?

CIECAM97 is one of many color appearance models. It was developed last year as an international standard color appearance model that borrows the best ideas from many different color appearance models. CIECAM97 currently is only developed for a single input color and a certain illumination environment. It breaks down the illumination environment and the color stimulus in the following way:
1. Color Stimulus: The color stimulus is the color being observed. It is defined to cover about 2 degrees of visual angle.
2. Background: The background is the area immediately surrounding the stimulus. It extends from the edges of the stimulus to about 10 degrees of visual angle in all directions.
3. Surround Field: The surround field extends to the limit of vision from the background. CIECAM97 defines a surround field as average, dark, dim, etc.

The CIECAM97 model uses the color stimulus, the color of white, the background, and the surround field to calculate its representation of the stimulus color in the viewing condition specified. CIECAM97 represents the stimulus color using lightness, chroma and hue which are described below:

• Lightness: Lightness represents the general illumination of the color. It is related to what people usually refer to as brightness. A color on a black background will have a greater lightness value than the same color on a white background.
• Chroma: Chroma or colorfulness, represents how rich a color appears. A color that is close to a gray value has a low chroma, while a deep rich color has a high chroma or colorfulness. Under an overcast sky, objects usually have a low chroma. When the sun comes out, colors become much more vivid and thus have a higher chroma.
• Hue: Hue represents the actual color one observes: yellow, blue, red, etc. A fruit as it ripens undergoes a change in hue: transitions from green to orange or red.

Once lightness, chroma and hue information is computed for the color stimulus and the viewing conditions, it can be used to tranform the color to a different viewing condition. The ultimate goal of the CIECAM97 model is for the two colors to appear exactly the same.

CIECAM97 Specifics

Input Data
• XYZ: Relative tristimulus values of color stimulus
• La: Luminance of the adapting field (cd/m*m)
• XwYwZw: Relative tristimulus values of white
• Yb: Relative luminance of the background

Viewing Condition Parameters

• c: Impact of surround
• Nc: Chromatic induction factor
• FLL: Lightness contrast factor
• F: Degree of adaptation factor

Output Data

• J: Lightness: Overall illumination of the color, related to brightness
• C: Chroma: Richness of the color, related to colorfulness
• h: Hue: Classifies the color as an angle, in degrees, around a color wheel