摘要
运用模式识别技术的SIMCA法对不同产地的45种山药样品进行了道地性与非道地性的模式识别方法学研究。凭借红外光谱具有的指纹特性,构建山药样品的红外指纹图谱,作为模式识别提取的特征数据,以随机分取的22个样品为训练集,剩余23个样品为试验集,该方法的正确识别率为70％,取得了满意的分类效果。
To identify the origin of tuber dioscoreae, 45 samples were studied by soft independent modeling of class analogy (SIMCA) in this paper. The combination of Fourier transform infrared spectroscopy (FTIR) with mathematic method was used to classify the trueborn and non-trueborn samples. The samples were chosen randomly as modeling group and predicting group. The correctness of classification was 70%. This approach was proved to be a reliable and practicable method for trueborn quality analysis of tuber dioscoreae.
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
2003年第2期258-261,共4页
Spectroscopy and Spectral Analysis
基金
河南省科委自然科学基金(0111023000)
