Dry colour prediction of leather from its wet state

The present investigation describes the development of a regression model for dry colour prediction based on the colour of wet leather. Such a prediction will be useful in the colour matching of leather – the time delay in drying of wet leather and determination of the final colour can be avoided. The use of computer technology interfaced with a reflectance spectrophotometer provides a novel approach to the prediction of dry colour from wet colour. In this study, five different regression models, i.e. linear fit, log fit, quadratic fit, log quadratic fit and cubic fit, were developed, with the dry reflectance of the leather sample as a dependent variable and the wet reflectance of the leather as an independent variable. The relationship between moisture content and surface leather colour, a prerequisite for enabling prediction, was studied. The obtained log quadratic model can be used for accurate prediction of dry colour from wet colour.