Display Neutral Color in Various Viewing Conditions

Abstract

The human visual system adjusts its sensitivity to maintain an object's color appearance across various viewing conditions. Unlike surface colors, which change spectrum depending on ambient illuminant, displays emit consistent light. As our eyes adapt to ambient conditions, the color appearance of the display changes. Therefore, adjusting the display white point according to the ambient lighting is crucial for maintaining color consistency and achieving high-quality display images. In this study, the display neutral color, which is the color that is perceived as having no hue, is investigated under various ambient conditions. To estimate the exact chromaticity of the neutral color, hue proportion data is collected using the double forced-choice method. The neutral color is estimated as the mean value of the Gaussian function fitted to the neutralness based on the chromaticity of stimuli. Neutral color data is collected in 77 sessions, divided into two groups based on objectives. The 67 sessions in the first group focus on the effect of ambient chromaticity, ambient luminance, and display luminance on the display neutral color. The 10 sessions in the second group evaluate the effects of experimental settings, such as background and surrounding colors, the presence of a display image with a different white point from the ambient chromaticity, the presence of a reference white tile, and the size of the display on the display neutral color. The experimental results demonstrate that the display neutral color is primarily influenced by ambient lighting, and observers adapt to the absolute intensity of the ambient illuminant irrespective of the reflected intensity in the experimental environment. As the correlated color temperature (CCT) of the ambient lighting increases, the CCT of the display neutral color also increases. When the ambient chromaticity remains identical, changes in the neutral color occur based on the surround ratio rather than the absolute intensity of the display and ambient illuminant. The surround ratio is defined as the ratio of the luminance of the white tile to the display luminance. In conditions where the surround ratio is zero, indicating a darkroom environment, the neutral color remains approximately 7,200 K, regardless of the display luminance. As the surround ratio increases, the display neutral color gradually shifts toward the ambient chromaticity. Furthermore, when the surround ratio exceeds 1.0, it converges to a specific chromaticity defined as the adapted white, which varies depending on the ambient chromaticity. Since existing chromatic adaptation transforms (CATs) have limitations in predicting display neutral colors, a new model, the Neutral Color Prediction (NCP) model, is proposed. It predicts display neutral colors based on the display luminance, given the ambient chromaticity and intensity. The cone responses of the neutral color are predicted between those of darkroom white and adapted white using the power function of the surround ratio. The NCP model shows good performance in predicting the preferred white point of images with content. Additionally, it works in different forms of display viewing conditions, such as virtual reality and see-through displays.