An algorithm for recalibrating a color monitor's RGB input‐output relations is presented that requires only a single measurement of a properly chosen reference stimulus. For the application under concern, i.e., reproduction of 35 different colored patches that were used as stimuli for psychophysical experiments on color constancy, the reference stimulus was a white (D65) presented at a luminance corresponding to the mean of the test stimuli. Three sets of data were obtained for evaluating the algorithm's error reduction power for a given stimulus configuration. These relate to different ways in which the monitor can get out of calibration. That is, slow, but cummulative changes over time, fast changes dur to gun interaction (resulting from changed stimulus conditions), and error introduced by a different setting of the monitor's brightness control. Additional experiments were performed to evaluate the effect of background intensity and color. The algorithm was found to be quite effective in dealing with the instantaneous changes (gun interaction, brightness control), and also for keeping track of the slow changes that may finally necessitate a full recalibration of the monitor. Copyright © 1990 Wiley Periodicals, Inc., A Wiley Company
