油菜籽品种的红外光谱鉴别研究

Identification of Rapeseed Varieties by Infrared Spectroscopy Combined with Spectral Retrieval

用傅里叶变换红外光谱（FTIR）技术结合光谱检索的方法对油菜品种进行鉴别研究。测试了15个油菜品种75份菜籽样品的红外光谱，利用光谱软件Omnic8．0建立了两个光谱数据库Lib01和Lib02，Lib01仅云南的9个品种的平均光谱组成，Lib02由全部15个品种的平均光谱组成。未知样品光谱分别与光谱数据库Lib01进行全谱范围的相关性检索和平方微分差检索，相关性检索的正确率达到77．8％，平方微分差检索的正确率达到82．2％。未知样品光谱分别与光谱数据库Lib02进行全谱范围和指定在1700～950c盯。光谱范围的平方微分差检索，全谱范围的检索正确率达到82．7％，指定光谱范围的检索正确率达到90．6％。研究结果表明，FTIR结合光谱检索的方法有望成为一种简便、易行的可以鉴别油菜籽品种的方法。

The infrared spectra of 75 samples of 15 rapeseed varieties were obtained by means of Fourier transform infrared （FT IR） spectroscopy technique. The results show that the infrared spectra of the samples are very similar. By the spectral profes sional software Omnic8. 0, two spectral databases were constructed. Lib01 includes the average spectra of 9 samples from Yun nan, while Lib02 is constructed from the average spectra of all 15 varieties. The correlation search and the squared differential difference retrieval of the spectra of Yunnan samples were performed with the spectral database Lib01 in full spectrum range, re spectively. The obtained matching values reflect that the samples have the closest correlation with each other only within the same species, with the correct rate of 77. 8% for the correlation search, and that of 82. 2% for the latter. The squared differenti al difference retrieval for the 15 sample spectra was carried out with Lib02 in full spectral range and in the specified range of 1 700～950 cm-1 , yielding correct rate of 82.7% and 90.6% respectively. These results show that FTIR combined with the spectral retrieval method can identify the different varieties of rapeseed, and the squared differential difference retrieval in the specified spectral range is a more effective and simple way for distinguishing different varieties of rapeseed.