基于PCR和PLS的沙棘汁品种近红外光谱研究  

Research on Varieties of Sea Buckthorn Juice by Near-Infrared Diffuse Reflectance Spectroscopy Based on the PCR and PLS

作　　者：张海红[1] 张淑娟[1] 介邓飞[1] 赵华民[1] 

机构地区：[1]山西农业大学工程技术学院,山西太谷030801

出　　处：《山西农业大学学报（自然科学版）》2010年第1期46-48,共3页Journal of Shanxi Agricultural University(Natural Science Edition)

基　　金：山西省科技攻关项目(2007031109-2)

摘　　要：建立可见-近红外漫反射光谱与沙棘汁品种之间的数学模型,以评价可见-近红外漫反射光谱技术快速检测沙棘汁品种。采用美国ASD公司的FieldSpec3光谱仪对三种不同品种的沙棘汁进行光谱分析,各获取30个样本的光谱数据,对原始光谱进行一阶微分和二阶微分预处理,并利用偏最小二乘法(PLS)数学校正方法对三种不同预处理的光谱数据建模。结果表明,采用二阶微分预处理数据,应用PLS方法建模较好,其校正模型相关系数为0.9992,均方根误差为0.0317;采用主成分回归(PCR)和偏最小二乘法(PLS),对沙棘汁的二阶微分数据进行分析比较,结果也表明,基于二阶微分数据,应用PLS方法建模较为理想,其预测集的相关系数为0.9988,所测预测样本的均方根误差为0.0392。近红外光谱可作为一种快速、有效的无损检测方法来识别沙棘汁的品种。The objectives of this study are to establish mathematical relationship between visible and near-infrared （Vis NIR） diffuse reflectance spectroscopy and sea buckthorn juice varieties, and to evaluate the applicability of Vis_NIR spectroscopy technique for fast measurement of the sea buckthorn juice varieties. A Fieldspec3 spectroradio meter was used for collecting 30 samples spectra data of the three kinds of sea buckthorn juices, Then the first and second derivatives were calculated using Vis_NIR diffuse reflectance spectroscopy, and principal component regression （PCR） and partial least square （PLS） regression were used to establish mathematical models to analyze the spectral data with three pretreatments The best prediction results were obtained, which was based on second derivative with PLS model, and its correlation coefficients of calibration set was 0. 9992 , and root mean standard error of correction （RMSEC） was 0. 0317. Based on the second derivative with PLS model, and its correlation coefficient of prediction set is 0. 9988 ,and root mean standard error of correction （RMSEC） is 0. 0392. Vis_NIR spectroscopy is a fast and available method for non-destructive detection of Sea buckthorn Juice Varieties.

关 键 词：可见-近红外漫反射光谱 快速检测 沙棘汁品种 PCR PLS 

分 类 号：O433.4[机械工程—光学工程]