This paper evaluates museum lighting sources by proposing indices of color variety perceivable in artworks under illumination. We adopt three types of white LED with high Ra values of 90, 96 (excited by blue ray, B-LED), and 97 (excited by purple ray, P-LED) where correlated color temperatures (CCTs) are 3000K and 4000K, and we use some other supplementary LEDs. These LEDs are applied to two groups of oil paintings. The first depicts the same hydrangea motif drawn under two different lightings, and the second is two masterpieces in the National Museum of Western Art, Tokyo. We propose and test three indices of color variety. Two are color standard deviations σ³₇₆ and σ³₀₀ of CIELAB color space measured by CIE1976 color difference ΔE*_ab and CIEDE2000 color difference ΔE₀₀, respectively. The third index is color diversity S obtained via color entropy of an artwork’s probability distribution in the CIELAB color space discretized with ΔE₀₀. A color reference chart and a numerical model of color distribution are preliminarily examined to show basic properties of the indices. Experimental results of the paintings reveal the effectiveness of σ³₀₀ and S and show that P-LEDs may yield higher color variety than B-LEDs with Ra values nearly equivalent to those of P-LEDs with considerable certainty. We discuss similarity between the properties of σ³₀₀ and those of S from the viewpoint of distribution bias in the CIELAB color space and mention a kind of correlation between the similarity and Ra value.

In the field of tunnel lighting design, a linear relationship in double logarithmic plots with a slope of minus 0.5 has been widely referred to as the relation between average road surface luminance and the overall uniformity of road surface luminance with keeping object visibility constant. This indicates that if the uniformity of road surface luminance is increased, the average road surface luminance could be lowered while maintaining object visibility at a certain level. Advancements in lighting technology, such as LED, have enabled highly uniform road surface luminance; however, lowering the average road surface luminance is a concern given the increase in elderly drivers. In this study, the relation between average road surface luminance and the uniformity of road surface luminance was investigated for young and elderly observers over a range of low to high uniformity. Our results indicate a non-linear relation in double logarithmic plots, especially in the low road surface luminance and high uniformity region. Furthermore, a marked increase in average road surface luminance is necessary for elderly observers to maintain the same visibility level as that of young observers.

Due to the prevalence of infectious diseases caused by viruses and bacteria, UV sterilization has been attracting attention. Since UV is invisible, the repeated prototyping, measurement, and sterilization testing involved in the product development is both time-consuming and costly. This paper discusses the effect of the light receiving surface spacing on the results of the analysis of UV integrated light exposure, and shows that the error increases as the light receiving surface spacing increases. We calculated the UV integrated light exposure of the water sterilizer, and found that, based on a 5 mm light receiving surface spacing, the error was 0.17% when the light receiving surface spacing was 100 mm and the reflectance of the component was 90%, and the error was 6.03% when the reflectance of the component was 30%. Next, using the average of the irradiance at three locations on the light receiving surface to calculate the UV integrated light exposure, we found that the maximum error was 0.51% when the reflectance of the component was 90%, and the maximum error was 23.89% when the reflectance of the component was 30%.We therefore recommend examining the analysis results by setting multiple patterns of light receiving surface spacing for the analytical model and performing optical simulations with light receiving surface spacing where the errors in the analytical values and the computational resources are acceptable. In addition, we describe the process of applying this analysis method to the development of UV sterilization equipment.

We have designed and evaluated a spectral distribution of an illumination light source that predicts color visibility in mesopic vision by applying a color recognition characteristic model that is based on the opponent color response function we previously proposed. We also designed a spectral distribution of illumination light sources to reduce the change in color visibility in response to changes in illuminance level. Subjective experiments were conducted using the designed illumination light source, and the results showed that an illumination light sources light source can be realized that reduces the change in color visibility in response to changes in illuminance level.