贵定云雾茶近红外识别模型的初步建立.pdfVIP

φ (j)，tt营 l呈报 2013 ,29(3):215-220 Chinese Agricultura1 Science Bulletin 贵定云雾茶近红外识别模型的初步建立 袁鑫l 气谭红2 ，杨鸿波2 ，何锦林2 ，董二慧 IA ，马凯口 ( I 贵州大学化学与化工学院，贵阳 550025; 贵州省理化测试分析研究中心，贵阳 550002) 摘要:为了快速、准确的鉴定贵定云雾茶的真伪，以贵州省各地茶叶为研究对象，应用近红外漫反射光 谱技术对贵定云雾茶进行定性分析。结合判别分析法(discriminant analysis , DA) ，光谱范围选取 I 4235.35-4252.46、 5484.32 -5497.04、 5999.06~6310.39 cm- ，并采用 SNV 结合一阶导数、Norris 平滑的光 谱预处理方式，主成分因子数选5 ，建立识别模型。结果表明，模型的正确识别率达到88.6%。为快速准 确识别茶叶提供了一种新思路。 关键词:近红外光谱;茶叶;识别模型;品质成分 中图分类号:0657.3 文献标志码:A 论文编号:2012-1934 Preliminary Construction of NIR R配ognition Model for Guiding Fog Tea l 2 l l 2 Yuan Xin .\ Tan Hong , Yang Hongbo\ He linlin\ Dong Erhui Ma Kai • 气 (School 01 Chemistry and Chemical Engi TU!ering, Guizhou Univers町， Guiyang 550025; Guizhou Research Center 01Physical Test and Chemical Analys白， Guiyang 55创)()2) Abstract: This paper viewed the tea of Guizhou Province as research object, and attempted to establish rapid qualitative analysis method for identification the authenticity of Guiding fog tea by NIR spectroscopy. The results indicated that: using discriminant analysis , spectraI wavelength was 4235.35-4252 .46, 5484.32-5497.04, 5999.06-6310.39 cm-\ spectral data preprocessing method was SNV, First Derivative and norris smooth , the factor was 5, recognition rate of the model was 88.6%. A new idea by the quick and precise identification of tea was offered in this paper. Key words: near-infrared spectroscopy; tea; recognition model; quality component