In this work, linear and exponential weighted principal component analysis techniques based on spectral similarity were employed for the prediction of dye concentration in coloured fabrics, which had been dyed with three component dye mixtures. The matching strategy was based on the equalisation of the first three principal component coordinates of the weighted reflectance curves of the predicted and target sample in a dynamic 3D eigenvector space. The performance of the proposed algorithm was evaluated by the root mean square differences of the reflectance curves and the relative error of the concentration prediction, as well as the metamerism index. The obtained results indicated that the developed exponential weighted principal component analysis method is more accurate than the spectrophotometric method and the simple principal component analysis matching strategy.
There are several approaches to dyeing recipe formulation in the textile industry, because the accuracy of computer recipe prediction is not very high. There are several theories such as colorimetric and spectrophotometeric appraches on color reception. This work addressed the effect of the calibration method as the calculation of unit k/s and spectral overlapping of the components of the mixture on the accuracy of recipe prediction in dyeing nylon with acid dye. To this aim, the unit k/s was calculated using single and binary mixture samples. Two types unit k/s were used for the recipe prediction of ternary mixture dyed samples with low and high overlapping reflectance spectra. The obtained results indicated that the unit k/s of single component differed from the unit k/s of bicomponent mixture. The accuracy of recipe prediction using unit k/s of single component was less than that of the binary mixture. However, the performance of the recipe prediction of high level overlapping was found to be more than that of the low level overlapping mixture. Subsequently, the accuracy of color matching depended on the reflectance spectra similarity (spectral overlapping) of the components of the mixture.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.