Abstract:Geographical origin of medical herbs is an important factor of the quality of many traditional Chinese herbal medicines. The objective of this study is to investigate whether FTIR spectroscopy coupled with pattern recognition techniques could effectively discriminate geographical origins of medical herbs. Nearest neighbor method (NNM) and a SVM-based multiclass classifier were employed to discriminate 269 angelicae dahuricae radix (ADR) samples from 4 provinces in China and 380 salviae miltiorrhizae radix (SMR) samples from 6 provinces. A leave-one-out cross-validation accuracy of 99% was achieved by the multiclass classifier. The study shows this classification scheme can be a highly accurate approach for the discrimination of medical herbs of different origins.
Key words:FTIR spectrum;Pattern recognition;Chinese medical herb;Support vector mechanism
刘沭华1,张学工1*,周 群2,孙素琴2 . 模式识别和红外光谱法相结合鉴别中药材产地[J]. 光谱学与光谱分析, 2005, 25(06): 878-881.
LIU Shu-hua1, ZHANG Xue-gong1*, ZHOU Qun2,SUN Su-qin2 . Use of FTIR and Pattern Recognition to Determine Geographical Origins of Chinese Medical Herbs . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2005, 25(06): 878-881.