''Visual clarity'' or ''feeling of contrast'' of object colors under illumination is affected significantly by changing color-rendering property of light source used. Though the feeling of contrast is considered one of the most important characteristics on color-rendering properties of light sources, it cannot be estimated adequately by using the present R a method. The new index FCI is proposed for estimating the effect of feeling of contrast quantitatively under any light sources. The FCI is derived using a simple transformation of the gamut area, which is constituted by a specially selected fourcolor combination in CIE LAB color space. The FCI correlates well with the illuminance ratio for equal feeling of contrast (or equal visual clarity) on various light sources reported so far. Using the FCI together with the present CIE R a , the color-rendering capability of a light source can be well clarified.

The visual clarity of a lighting environment is sign$cantly affected by changing the general color rendering index (Ra) of its illumination. This eflect has been studied by a number of researchers, but the cause of this effect has not been thoroughly studied. In order to clarify this, the mutual relations between visual clarity, lightness perception, and feeling of contrast are analyzed by using object colors under illuminations with various Ra values. The results obtained are as follows. (1) The visual clarity of a lighting environment is diflerent .from the lightness perception of the object colors in the environment. (2) The change in visual clarity of a lighting environment is highly correlated to a feeling of contrast between object colors under the illumination. (3) The effect of visual clarify is estimated eflectively by assessing the feeling of contrast using a specially selected four-color combination. (4) The illuminance for equal visual clarity for any illumination is predicted by the equality offeeling of contrast under the same illumination specijied by the gamut area made by the component colors of the four-color combination in a brightness and colo~uIness space. (5) The effect of visual clarity under various illuminations cannot be predicted by using their Ra values. o 1994 John WiIcy & Sons, Inc.

A formulation for predicting the Helmholtz‐Kohlrausch (H‐K) effect is given to the new VCC (variable chromatic color) method, which can derive the luminance factors of various chromatic object colors that match in perceived lightness with a fixed achromatic color. By using the formulation, the H‐K effect by the VCC method is predicted within the whole chromaticity gamut including spectral colors. the magnitude of the effect in logarithmic scale by the VCC method is estimated to be twice as large as that by the VAC (variable achromatic color) method previously used. It is also stated that the VCC method is practically more important than the VAC method. the effectiveness of the new formulation is confirmed by comparing its predictions with the observations in two kinds of experiments. © 1994 John Wiley & Sons. Inc.

In the representation of the Helmholtz‐Kohlrausch effect, a new method (the VCC method using variable chromatic colors) is proposed in addition to the method used so far (the VAC method using variable achromatic colors). The new VCC method showed a significantly larger Helmholtz‐Kohlrausch effect than the VAC method previously used. For practical applications, the results by the VCC method are more effective than the VAC method. Three kinds of experiments are reported for clarifying the difference in effect between the two methods. © 1994 John Wiley & Sons, Inc.

The nonlinear color‐appearance model of Nayatani et al. is reformulated by changing the fundamentals from the Pitt to the Estévez‐Hunt‐Pointer primaries. The predictions of the reformulated model correlate with the Munsell and the NCS schemes better than those of the original. Formulas for various color‐appearance metrics are proposed, such as colorfulness, chroma, saturation, brightness (including a chromatic‐component contribution), and others. In the present model, the normalizing illuminant D65 is used in the transformation from the CIE tristimulus values to those in the fundamental‐primary system, and the normalizing illuminance is kept at 3,000 lx. The effects of changing the normalizing illuminant to D50 and the normalizing illuminance to 1,000 lx are studied. The results, however, show insignificant differences in the predictions of color appearance.