摘要
【目的】采用快速可靠的方法对薰衣草品种进行鉴别。【方法】采用傅里叶变换红外光谱法结合主成分分析(PCA)法及偏最小二乘判别(DPLS)法对不同品种薰衣草进行鉴别分析。【结果】对不同品种薰衣草的快速FTIR光谱鉴别,采用PCA法对光谱数据进行聚类分析,得到四种不同品种薰衣草的特征差异。用特征谱图作为输入建立判别偏最小二乘模型,四个品种薰衣草共47个分别建立偏最小二乘判别(DPLS)模型,对未知的16个样本进行预测,DPLS模型的校正决定系数(R2)为0.991 0,交叉验证均方根标准差(RMSEE)是0.119 3,预测均方根标准差(RMSEP)为0.146 2,品种识别率为100%。【结论】PCA对不同薰衣草具有较好的聚类作用,DPLS模型能对薰衣草的品种进行较好的识别,为薰衣草品种快速鉴别提供了较好的分析方法。
[ Objective] The aim of this study was to provide a fast and reliable scientific approach to identify different varieties of lavenders. [ Method ] Different varieties of lavenders were analyzed by Fourier transformation infrared spectroscopy （F-FIR） combined with the principal component analysis （PCA） and partial least - squares discriminate （DPLS）. [ Result ] PCA had very good clustering effect on different types of lavenders. The characteristic band spectrum was used to build DPLS. Forty - seven lavender of four varieties were used to build DPLS and the unknown 16 samples were predicted by the models DPLS model calibration determination coefficient （R2） is 0. 991 0, cross validation root mean square standard deviation （RMSEE） was 0.119 3, and forecast root mean square standard deviation （RMSEP） is 0. 146 2, The breed recognition rate is 100%. [ Conclusion] Lavenders were analyzed by principal component analysis （PCA）, which could intuitively distinguish lavenders varieties. DPLS model provided good recognition for the varieties identification of lavenders.
出处
《新疆农业科学》
CAS
CSCD
北大核心
2013年第4期614-619,共6页
Xinjiang Agricultural Sciences
基金
国家自然科学基金(21265021)
关键词
薰衣草
傅里叶变换红外光谱
偏最小二乘判别
主成分分析
lavenders
Fourier transformation infrared spectroscopy
discriminate
principal component analysis partial least - squares
