基于GABP算法的油墨配色研究  

Ink color matching research based on GABP algorithm

作　　者：张阁 郑元林 廖开阳 刘梦莹 王晓彤 ZHANG Ge;ZHENG Yuanlin;LIAO Kaiyang;LIU Mengying;WANG Xiaotong(School of Printing, Packaging Engineering and Digital Media Technology, Xi’an University of Technology, Xi’an 710048, China;Key Laboratory of Printing and Packaging Engineering of Shaanxi Province, Xi’an 710048, China)

机构地区：[1]西安理工大学印刷包装与数字媒体学院,陕西西安710048 [2]陕西省印刷包装工程重点实验室,陕西西安710048

出　　处：《西安理工大学学报》2019年第1期113-119,共7页Journal of Xi'an University of Technology

基　　金：陕西省教育厅科研计划资助项目(17JK0990)

摘　　要：为了提高印刷油墨配色精度,本文提出了使用遗传算法(GA)优化的BP神经网络模型GABP。首先采集了颜色样本的光谱数据和CMYK四色网点面积率分别作为输入值和输出值,然后通过遗传算法优化了BP神经网络的结构和参数,并进行了训练,最后将模型的预测精度做对比分析,使用未经遗传算法优化的BP神经网络模型的平均误差为8.6%,使用GABP模型的平均误差为4.5%。结果表明,经遗传算法(GA)优化的油墨配色模型预测精度有了大幅度提高,对印刷企业油墨配色有很好的应用价值。In order to improve the color matching precision of printing ink,this paper proposes a genetic algorithm(GA)to optimize the BP neural network model.The spectral data and CMYK chromaticity values are collected.Based on the BP neural network,the network structure and parameters are optimized by the genetic algorithm,with the results given to the neural network for training.In this training,the spectral data is taken as the network input value and CMYK as the network output value.Finally,the average color difference of BP neural network is 8.6%,and the average color difference of GABP 4.5%.Compared with the BP neural network,the accuracy has been greatly improved,which is of excellent value for ink color printing business.

关 键 词：油墨配色 遗传算法 神经网络 

分 类 号：TS8[轻工技术与工程]