数码印花的光谱预测模型研究  

Research on Spectral Prediction Model Based on Textile Digital Printing

作　　者：马倩云 王强 

机构地区：[1]杭州电子科技大学数字媒体与艺术设计学院,浙江杭州310018

出　　处：《杭州电子科技大学学报（自然科学版）》2017年第6期67-72,共6页Journal of Hangzhou Dianzi University：Natural Sciences

基　　金：国家科技支撑计划课题资助项目(2012BAH91F03)

摘　　要：以色料减色法理论为基础,通过对数码印花的基色网点输入值与光谱反射率的非线性关系和光谱叠加特性的实验分析,提出基于光谱特征的基色光谱反射率函数及与叠印次数相关的调整因子,建立了数码印花的光谱预测模型.实验结果表明:提出的数码印花光谱预测模型的光谱精度和色度精度均优于YNSN模型和CYNSN模型.同时,弥补了YNSN模型和CYNSN模型预测时,因基色数量过多而导致的空间复杂度激增的缺陷,满足数码印花多基色的实际生产需求.A spectral prediction model for textile digital printing is proposed in this paper.Based on color subtractive mixture theory,the primary color spectral reflectance function of spectral characteristics and the relevant adjustment factor n related with the number of color are proposed during the research,which through analyzing the non-linear relationship between printing dot coverage input and ink spectral reflectance,and spectral superposition characteristic.The result shows that the proposed spectral prediction model is superior to YNSN and CYNSN in aspects of spectral accuracy and colorimetric accuracy.Furthermore,it overcomes the deficiencies of prediction by complexity of the dot overprint numbers in YNSN and CYNSN,and meets the requirements for multi-color textile digital printing.

关 键 词：数码印花 光谱预测 叠加特性缺失 叠印次数 

分 类 号：TS805.4[轻工技术与工程] TS194