文献类型：期刊文献

中文题名：近红外漫反射光谱法测定油茶籽含油量的研究

英文题名：Determination of Oil Content of Camellia oleifera Abel. Seeds by Near Infrared Reflectance Spectroscopy

作者：原姣姣[1] 王成章[1,2] 陈虹霞[1] 叶建中[1,2]

第一作者：原姣姣

机构：[1]中国林业科学研究院林产化学工业研究所生物质化学利用国家工程实验室国家林业局林产化学工程重点开放性实验室江苏省生物质能源与材料重点实验室油茶科学中心加工利用实验室;[2]中国林业科学研究院林业新技术研究所

年份：2011

卷号：31

期号：3

起止页码：28-32

中文期刊名：林产化学与工业

外文期刊名：Chemistry and Industry of Forest Products

收录：CSTPCD;;Scopus;北大核心:【北大核心2008】；CSCD:【CSCD_E2011_2012】；

基金："十一五"国家科技支撑计划资助(2009BADB1B03);中央级公益性科研院所基本科研业务费专项资金(CAFYBB2008036);浙江省省院合作项目(2009SY01)

语种：中文

中文关键词：近红外漫反射;油茶;含油量

外文关键词：near infrared reflectance spectroscopy; Camellia oleifera Abel.; oil content

分类号：TQ351.0;O657.3

摘要：以索氏提取法分析的30份油茶籽的含油量值为化学法测定值,结合相应的近红外漫反射光谱,通过多种预处理方法和回归方法建立了较精准的分析模型。结果表明：所建模型经＂一阶导数＋SG＋MSC＂预处理组合最好,校正相关系数（RC）和校正标准误差（RMSEC）分别为0.925 67和2.53;交叉检验相关系数（RCV）和交叉检验校准误差（RMSECV）为0.917 27和2.67;最优波段为4200.20-4088.35 cm^-1和4666.89-4639.89 cm^-1。还比较了偏最小二乘法（PLS）和主成分回归（PCR）方法建模结果,以偏最小二乘法回归建模效果较好。10份验证样品的预测值和化学法测定值的相关系数（r）为0.897 8,预测效果较好。
Near-infrared reflectance spectroscopy of Camellia oleifera seed was used to determine its oil content,which was compared with the oil content data of 30 samples of C.oleifera Abel.seeds obtained by conventional Soxhlet extraction method.Combined with the corresponding near infrared reflectance spectroscopy,the accurate model was well established with different pretreatment methods and regression analysis.The results demonstrated that the correlation coefficients of calibration（RC） and the root mean square errors of calibration（RMSEC） through first derivative、multiplicative signal correction（MSC）and Savitzky-Golay filter spectral pretreatment methods were 0.925 67 and 2.53 respectively.The correlation coefficients of cross-validation（RCV） and the root mean square errors of cross-validation（RMSECV） were 0.917 27 and 2.67 respectively.The best spectral ranges were 4200.20-4088.35cm^-1 and 4666.89-4639.89cm^-1.Furthermore,the comparison of the calibration models by PLS and PCR showed that PLS was the optimal regression method.The correlation coefficients（r） between NIRS value and chemical value was 0.8978.All parameters indicated that near infrared reflectance spectra information of C.oleifera seeds could be used to predict the oil content fast and exactly.