A STUDY OF THE JUDGEMENT TO STAIN-FASTNESS OF DYEINGS BY STATISTICAL METHOD

Abstract

The staining rating number estimated by Gray Scale was analyzed by the use of successive categories theory in order to give objective validity to the judgement on the stain-fastness of dyeings. The rating number of stained cloth specimens, which were rubbed on black leathers, was judged twice according to Gray Scale by nineteen technical experts. The visual judged data, which was an ordinal scale, were converted to an interval scale by the successive categories to determine the ranges of each Gray Scale steps. The color difference, ΔE, were correlated with the visual interval scale converted. The relationship between the visual interval scale and the color difference was calculated by the use of regression analysis and is expressed in the following form where _??_ is an estimated value from each Gray Scale steps on the visual interval scale, C is a constant when the correlation coefficient between S and exp(C×ΔE) becomes the maximum, and A and B are the regression coefficients, respectively. The correlation coefficient of 0.999 was obtained. The ranges between each Gray Scale steps were determined on the ΔE scale by the use of the above equation. It should be noted that these ranges enable the judgement on fastness of dyeings to give objective validity. The visual interval scale showed a non-linear relationship to Munsell value scale due to the difference in the quality of stimulus detected between both scales. That is, the Munsell value scale depends on the luminous reflectance, on the other hand, the visual interval scale is attributed to the color difference between stained and unstained cloths. It was found that the visual interval scale was suitable for Weber-Fechner's law.