SUMMARY
The concept of graphic contrast and issues of digital images' contrast control are considered from the viewpoint of visual perception of graphic information. Roles of color and contrast as its carriers are compared. The priority of the second attribute is shown. However, while color is always visible by definition, contrast can be latent. Contrast enhancement is studied in the context of detection of latent information. Being caused by perception of compared colors, informative contrast is predetermined in terms of hue-saturation-brightness. For the purposes of contrast enhancement the color model HSY is entered into consideration, where the choice of hue and saturation from color model HSB/HSV; of brightness – from color model XYZ is substantiated. Each visual contrast is defined through appropriate material contrast and its visibility functions. The structure of total contrast is resulted. Minding the outlined structure, ways to control contrast are scheduled. Special cases of its realization, when total contrast degenerates to contrasts of hues, saturations or brightness, are analyzed. Algorithms, constructed according to the optically-based theoretical formulae, are tested in natural experiment with reference to each partial contrast. Along with the initial image, the results of its informative processing are presented at magnification of contrast-enhancing factor, demonstrating the transition from zero total contrast (required image fragment is latent) up to its 100% (required fragment is visible). The relation of visibility thresholds of hue-contrast, saturation-contrast and brightness-contrast for the actualized image is estimated. The purport of estimation reflects the fact that chromatic contrast enhancement can be more effective than brightness one. Moreover, be brightness (of required fragment) and brightness (of background) equal, then it would be problematic for regular contrast-enhancing toolkit of modern graphic software to reveal the latent information. While the announced tool ensures successful contrast enhancement. The new approach actually triples volume of the accessible information, received by means of contrast enhancement from the same initial image.