Color Match Prediction for Munsell System Using Artificial Neural Networks

This present investigation demonstrates the ability of back-propagation neural networks to indentify the combination of dyes in color-matching. The reflectance and fundamental color stimulus values served as input data and the known concentration of dyes were used for each sample. 405 samples of one data set (textile) were used to train and test the performance of NN. Also, Munsell data set (Munsell Color Solid and Munsell Renotation System), having different origin, used to attest the well-performance of NN in color match prediction.

Keywords:

Artificial Neural Network , Fundamental color stimulus , Color match prediction , Munsell system