近红外光谱在印刷品颜色检测中的应用

来源期刊:中南大学学报(自然科学版)2009年第6期

论文作者:管力明 胡更生 卢红伟 林剑

文章页码:1655 - 1659

关键词:印刷品;近红外光谱;颜色检测

Key words:printing; near infrared spectroscopy; color detection

摘    要:为了对印刷品颜色进行快速、准确检测,应用近红外光谱技术(NIR)并结合偏最小二乘法(PLS)建立印刷品颜色检测模型。对近红外光谱获取的144个样本光谱曲线,应用主成分分析方法进行降维,维数为5。选取的主成分作为光谱优化特征子集以替代原来复杂的光谱数据。随后,将144个样本数据随机分为定标集和预测集,利用偏最小二乘法在103个定标集样本数据基础上建立印刷品颜色预测模型,应用此模型对41个预测集样本颜色进行预测。研究结果表明:利用PLS模型得到样本的实测值和预测值之间的预测决定系数(R2)为99.74%,预测平均相对误差为0.636%,表明利用近红外光谱技术检测印刷品颜色是可行的。

Abstract: In order to detect printing color fast and exactly, near infrared (NIR) spectroscopy technique combined with partial least square (PLS) was used to build the prediction model of color. NIR spectroscopy technique is a nondestructive, fast and accurate technique for the measurement of chemical components based on overtone and combination bands of specific functional groups. The pivotal step for spectroscopy technique is extracting quantitative data from mass spectral data and eliminating spectral interferences. PLS is used for the spectroscopic analysis. Firstly, the near infrared spectra of 144 samples are obtained, and then PLS is applied to reduce the dimension of the original spectra. The 144 samples are randomly separated into calibration set and validation set. PLS is used to build prediction model of chroma based on the calibration set, then this model is employed for the prediction of the validation set. Correlation coefficient (R2) of prediction and root mean square error prediction (RMSEP) are used as the evaluation standards. The results show that R2 and RMSEP for the prediction of chroma are 99.74% and 0.636%, respectively. Hence, PLS model with high prediction precision can be applied to the determination of chroma.

基金信息:国家自然科学基金资助项目
浙江省科技计划项目

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号