Laser-engraved color properties on cotton-spandex fabric predicted by artificial neural network

An artificial neural network (ANN) model was used for prediction of color properties, including color yield (K/S value) and CIELAB values, of laser engraved cotton-spandex fabrics using different combinations of laser processing parameters. Factors in the ANN model included fiber composition, fabric density, fabric mass, fabric thickness, and linear density of yarn. The ANN model produced superior results in prediction of color properties of laser engraved cotton-spandex fabrics when compared with a linear regression model. Factors influencing color properties were also investigated. Laser processing parameters played an important role in affecting the color fading effect of the treated fabrics. However, if fabric parameters were taken into consideration, fabric thickness dominated the color fading effect.