近红外漫反射光谱技术对小米产地溯源的研究  

Study on Traceability of Millet Origin by Near Infrared Diffuse Reflectance Spectroscopy

作　　者：宋雪健 钱丽丽[1] 周义[1] 于果[1] 于金池[1] 张东杰[1] 

机构地区：[1]黑龙江八一农垦大学食品学院,黑龙江大庆163319

出　　处：《食品研究与开发》2017年第11期134-139,共6页Food Research and Development

基　　金：黑龙江省教育厅科学技术研究项目资助(12541576);黑龙江省垦区科研项目(HKN125B-13-02);黑龙江省高等学校科技创新团队建设计划项目(2014TD006);黑龙江省应用技术研究与开发计划项目(GA14B104)

摘　　要：为建立小米产地溯源的快速检测技术,更好的维护地方名优小米品牌效益,试验利用近红外漫反射光谱技术对不同状态小米进行产地溯源鉴别,试验分别选取来自肇源和肇州两个小米主产区的144份小米样品,应用近红外漫反射光谱技术结合化学计量学对不同状态下的小米进行产地溯源研究,结果表明:在全波长范围内采用因子化法建立的定性分析模型和在特征波段范围内采用偏最小二乘法(PLS)建立的定量分析模型,对肇源、肇州两个小米主产区的小米籽粒和小米粉末的正确鉴别率均在90%以上,其中小米粉末的模型正确预测率要高于小米籽粒。因此,应用近红外漫反射光谱技术对不同状态小米产地溯源的鉴别具有一定的可行性。In order to establish a millet origin of the rapid detection technology, better maintenance of local fa- mous millet brand benefits, near infrared diffuse reflectance spectroscopy was used to identify the origin of mil- let in different states. A total of 144 millet samples from the main producing areas of Zhaoyuan and Zhaozhou were selected, and the near infrared diffuse reflectance spectroscopy combined with stoichiometry was used to study the origin of millet in different states. The results showed that the qualitative analysis model established by the factorization method in the whole wavelength range and the quantitative analysis model established by partial least squares （PLS） in the characteristic band range, and the correct identification rate of millet kernel and mil- let powder in Zhaoyuan and Zhaozhou two millet main producing areas was above 90 %, and the correct predic- tion rate of millet powder model was higher than that of millet kernel. Therefore, the application of near infrared diffuse reflectance spectroscopy to the different status of millet origin traceability of the rapid identification of a certain feasibility.

关 键 词：近红外漫反射光谱技术 小米 因子化法 偏最小二乘法(PLS) 产地溯源 

分 类 号：O657.33[理学—分析化学] TS210.7[理学—化学]