Tradeoffs in VDU Monitor Calibration and in Color Correction

Color calibration of a video display unit (VDU) determines the mapping from inputs (digital in the case of graphic VDUs and analog in the case of television) to tristimulus outputs. Assumptions are necessary to simplify this process. A universal assumption is spatial independence: that tristimulus values at a given monitor location are independent of the activations of other pixels on the monitor. Two other common assumptions are phosphor constancy (the relative spectrum of each phosphor is independent of its input) and phosphor independence (that the input to a phosphor influence the output of only that phosphor). In addition, assumptions are often made about the functional form of the phosphor luminance versus input gun voltage. The more of these assumptions one makes, the easier and more automatic can be the monitor calibration, but with a tradeoff against accuracy. Adjustments of the monitor are made during calibration to help ensure the invariance of the neutral chromaticity under scaling of the gun voltages by the same factor, to help match the gamut of the received colors to the gamut of the phosphors, and (in the case of digital inputs) to ensure adequate color resolution. These adjustments select a monitor white chromaticity. Color correction -- replete with its own assumptions -- then transforms a viewed scene from one illuminant to what it would appear under a light with the chromaticity of monitor white. The paper describes the implications and tradeoffs of all these assumptions.