E5 Color Optics and Chromatic Perception

Topic

As described by the photoreception model, each of the three types of retinal cones—S-cones, M-cones, and L-cones—responds with a different intensity to the same light, depending on the specific spectral sensitivity of that cone type. The color ultimately perceived does not depend on the light's full spectral composition, but solely on the combination of those three responses—from the S, M, and L cones—taken together. Consequently, this model treats color perception as a tristimulus model: it reduces any light—regardless of the complexity of its spectral composition—to just three numbers (one for each cone type), and it is this triplet of numbers that determines the perceived color.

The more similar the response triplets produced by two different lights in the three cone types are, the more similar the perceived colors will be; indeed, they are perceived as exactly the same color when the two response triplets are identical. This phenomenon is known as metamerism, and it occurs even when the two lights have completely different spectral compositions, because the visual system never compares those compositions directly; instead, it only compares the triplet of responses each light elicits in the S, M, and L cones. Thus, two metameric lights fool the eye in the same way, creating the perception of a single color where, physically, two distinct light combinations exist.

Since the perceived color is determined by just three numbers, it can be represented as a point within a color space—a three-dimensional space where each axis corresponds to one of the three cone responses. By separating brightness information from the response triplet and retaining only the proportions between the responses, this color space is reduced to a two-dimensional plane, and a color's position within that plane is defined by its chromatic coordinates. To ensure that different institutions and manufacturers can describe the same color in the same way, this model relies on a standard tristimulus model—internationally defined based on the average response of the three types of cones in a typical human observer—so that the resulting triplet of numbers and chromatic coordinates match regardless of who performs the measurement.

When two or more light sources combine and reach the eye simultaneously, the triplet of responses produced in the S, M, and L cones is the sum of the triplets that each light source would produce individually. Thanks to this additive property, the model allows for the prediction of the perceived color resulting from a mixture of lights without requiring knowledge of the mixture's full spectral composition, thereby enabling the simulation of color mixtures: one simply sums the triplets of the combined lights, response by response. This same principle is utilized by display screens to reproduce virtually any perceived color using just three primary lights in varying proportions, rather than reproducing the exact spectral composition that the color would possess in the real world.

When an eye lacks one of the three cone types—or when spectral sensitivity is altered—the triplet of responses produced by that eye in reaction to a given light source no longer matches the triplet produced by an eye with all three cone types intact; consequently, two lights that were previously distinguishable as different colors begin to be perceived as the same color. It is precisely this loss of distinction that the model for dyschromatopsia detects and classifies. For the same reason, any instrument that needs to measure or reproduce color reliably—from a display screen to a colorimeter—is calibrated against this tristimulus model by adjusting its sensors or primary light sources until the triplet of responses they produce matches the response that actual light would elicit in an eye with normally functioning S, M, and L cones.

ID:4313