Quantifying the perceptual translucency of the objects by spectrophotometric and imaging methods

Abstract

This study investigated the performance of various spectrophotometric methods in predicting visually perceived translucency and evaluated the efficiency of imaging techniques in quantifying translucency. We conducted the visual assessment in two stages using the pair comparison method. In the first stage, pairs of samples with identical colors but differing levels of translucency were compared. In the second stage, we compared pairs of samples with different colors but almost identical translucency. In the first stage, spectrophotometric methods exhibited high Pearson correlation coefficients, ranging from 0.96 to 0.99, with visual perceptions, demonstrating their accuracy in estimating translucency. Examination samples of different colors also revealed that among spectrophotometric methods, L*, absorption, and contrast ratio, with Pearson correlation coefficients of approximately 0.96, 0.96, and 0.98, respectively, had the strongest correlations with perceptual data. Using imaging techniques, the best result was obtained by comparing the lightness of the sample image on a black background to the same value on a white background, yielding a Pearson correlation coefficient of approximately 0.87. Additionally, we employed imaging-based methods for predicting translucency in real 3-D objects with simple shapes and limited colors. Despite the limitations, these methods produced promising results.