We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
基于BP神经网络的全色域彩色纱 颜色预测模型构建.
- Authors
李娟娟; 薛 元; 徐志武; 于 健; 曾德军
- Abstract
A color prediction model for full color gamut color spinning yarn is constructed by using the good nonlinear mapping ability of neural network to address the limitations of traditional colorimetric theory. A full color gamut model is constructed with gray, cyan, magenta and yellow as the four base colors, and 66 grid points in the full color gamut model are selected for spinning to prepare blended yarns and fabrics, and the spectral reflectance curves of 66 fabrics are measured by spectrophotometer, neural networks based on color prediction of blending ratio and blending ratio prediction of color values were established, and the accuracy of these two neural networks was evaluated according to the color difference value and mean squared error between the predicted and actual values. The results show that the color prediction model based on BP neural network can realize the nonlinear mapping between the color yarn color value and the proportion of the four base colors, and the average color difference of the test samples is around two. The subsequent plan is to using genetic algorithm or particle swarm optimization algorithm to further improve the accuracy of color prediction model for full color gamut colored yarn.
- Subjects
BLENDED yarn; SPECTRAL reflectance; PARTICLE swarm optimization; BLENDED textiles; GENETIC algorithms; SPUN yarns; SPECTROPHOTOMETERS; COLORIMETRIC analysis
- Publication
Basic Sciences Journal of Textile Universities / Fangzhi Gaoxiao Jichu Kexue Xuebao, 2023, Vol 36, Issue 4, p37
- ISSN
1006-8341
- Publication type
Article
- DOI
10.13338/j.issn.1006-8341.2023.04.006