Color classification and nuance reproduction for Burgundy Wines

Dr. Pierre Gouton
University of Burgundy, France


Abstract

In this work, we deal with the problem of colorimetric characterization, classification and reproduction of Burgundy wine. We propose to study the wines of Burgundy (France) that with the aim to represent their colors with limited number of hues (also called nuances). The study is based on spectral transmittance measures of a large number of wine samples. Then a quantization is performed on spectra obtained. Afterwards, the colors are classified in an appropriate color space (CIELAB) which provides good visual uniformity and thus offers the best discrimination between the different samples. To that end, we utilize an adaptive Kmeans. The preliminary results obtained are encouraging in the sense that they permit an accurate classification based on both spectral transmittances and hues, and allow for description and reproduction of the color of wines. In addition, we have we proposed a reproduction method base on a water color dye.