摘要
提出了应用模糊模式识别技术对来自不同产地的42个淫羊藿样品和40个人参样品的红外光谱数据进行处理,从而作出对此两类天然中药材的产地鉴别。用主成分分析法对上述两类药材的光谱变量作降维处理,然后用选取的主成分得分矩阵代替原始的光谱变量输入于模糊模式识别的模型中,从而求得上述药材每一个试样的贴近度和所属类别。模拟试验的结果表明:此两类天然中药材的产地鉴别的正确率可达92%以上。
The fuzzy pattern recognition was applied to origin identification of 42 samples of Epimedium macranthum and 40 samples of ginseng by IR spectrometry. Process of dimension reduction of the spectrum variables of these 2 groups of natural traditional Chinese medicines was carried out using the method of principal component analysis. Instead of the original spectrum variables the selected principal component score matrix was input into the model of fuzzy pattern recognition, to obtain the proximity and category for each of the samples. As shown by the results of simulation test, the rate of correctness of origin identification for the 2 kinds of natural traditional Chinese medicines attained to 92% and over.
出处
《理化检验（化学分册）》
CAS
CSCD
北大核心
2010年第7期778-780,783,共4页
Physical Testing and Chemical Analysis(Part B:Chemical Analysis)
基金
吉林省科技厅项目(20075020
200705C07)
国家科技支撑计划(2007BAI38B03)
吉林省教育厅十一五规划科研项目资助(吉教科合字[2008]第234号)
关键词
模糊模式识别
红外光谱
主成分分析
淫羊藿
人参
Fuzzy pattem recognition
IR-spectroscopy Principal component analysis Epirnedium macranthum Ginseng
