可见/近红外漫反射光谱无损检测南丰蜜桔维生素C的研究  

Nondestructive Measurement of Vitamin C in Nanfeng Tangerine by Visible/Near-Infrared Diffuse Reflectance Spectroscopy

作　　者：刘燕德[1] 陈兴苗[1] 孙旭东[1] 

机构地区：[1]江西农业大学工学院,江西南昌330045

出　　处：《光谱学与光谱分析》2008年第10期2318-2320,共3页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金项目(60468002;30560064);教育部新世纪优秀人才项目(NECT-2006-557);国家科技支撑计划(2006BADINA12-07);江西省井冈之星培养计划项目(2007-130);江西省自然科学基金项目(2007-187)资助

摘　　要：应用可见/近红外漫反射光谱对南丰蜜桔维生素C含量进行了无损检测研究。在谱区350～1800nm,应用主成分分析和偏最小二乘法对经过预处理光谱进行数学建模,主成分数和光谱区间选择分别由完全交互验证和回归系数法确定。定标模型对10个未知样品的预测结果是:预测相关系数为0.813,预测均方差为2.112mg(100g)-1,预测偏差为-0.810mg(100g)-1。文章表明利用可见/近红外漫反射光谱技术无损检测南丰蜜桔维生素C具有可行性。The present paper reports the study on visible/near-infrared （Vis/NIR） diffuse reflectance spectroscopy for nondestructive measurement of vitamin C in tangerine （Nanfeng, Jiangxi）. In the wavelength range of 350-1 800 nm, principal component analysis （PCA） and partial least squares （PLS） were used to build mathematic models with respect to pre-proeessing spectra. The principal components （PCs） and wavelength selection for calibration model were determined based on full cross validation and regression coefficients. The results, based on prediction for 10 unknown samples, are 0. 813, 2. 112 mg（100 g）^-1 and -0. 810 mg（100 g）^-1 for prediction correlation coefficient, root mean square error of prediction, and prediction deviation, respectively. The analysis methods and conclusions of this experiment will provide some reference values for non-destructive measurement of vitamin C and other quality indices in Nanfeng tangerine using Vis/NIR diffuse reflectance spectrometric technique.

关 键 词：可见/近红外漫反射光谱 无损检测 偏最小二乘法 维生素C 南丰蜜桔 

分 类 号：O657.3[理学—分析化学] TN247[理学—化学]