Colour recipe prediction in dyeing acrylic fabrics with fluorescent dyes using artificial neural network

Abstract

One of the most important processes in textile dyeing is the colour recipe prediction. Fluorescent dyes present difficulties for colour recipe prediction due to complex and non-linear mechanism of the dyes and dyeing process. For this reason, in this study, an artificial neural network technique is used to predict the colour recipes to match the referance colours of 100% acrylic knitted fabrics dyed by a combination of three basic dyes which are C.I. Basic Yellow 40, C.I. Basic Violet 39, and C.I. Basic Blue 3. C.I. Basic Yellow 40 and C.I. Basic Violet 39 are fluorescent and C.I. Basic Blue 3 has a brilliant blue shade. The neural network developed is a multilayer perceptron network: The inputs are the spectral radiance factors at wavelengths from 400 nm to 700 nm at 10 nm intervals and the target Outputs are dye concentrations. The results show that the artificial neural network is an appropriate and powerful tool for colour recipe prediction in dyeing acrylic fabrics with fluorescent dyes.