论文部分内容阅读
目的建立基于支持向量机的黄连饮片产地区分识别模型。方法采集4个产地多批黄连样本,量化外部特征值(包括形状、气味、味道等),并按《中国药典》2010年版方法测定内部特征值(包括水分、总灰分、醇浸出物,指标成分表小檗碱、黄连碱、巴马汀、小檗碱质量分数),在Matlab 7.0平台进行数据降维和融合,建立黄连产地区分模型。结果单独分析各项数据不能较好区分各产地黄连饮片,而采用支持向量机建模后所有特征叠加识别率达到97.1%,能准确区分各黄连饮片产地,内外特征的高识别率说明各特征子集间具有一定的互补性,可综合辨识不同产地黄连饮片的差异性。结论建立的基于支持向量机的识别模型,实现产地的区分,为黄连产地区分提供研究思路和基础。
OBJECTIVE To establish a discriminative model of the origin of Coptis chinensis slices based on support vector machine. Methods Four samples of Coptis chinensis from different producing areas were collected to quantify external characteristic values (including shape, odor, taste, etc.) and the internal characteristic values (including moisture, total ash, alcohol extract, Berberine, berberine, palmatine and berberine), and the data were reduced and merged on Matlab 7.0 platform to establish a discriminative model of Coptis origin. Results Individual analysis of each data can not distinguish between the different regions of Coptis chinensis Pieces. All the features have a 97.1% recognition rate after SVM modeling, and can accurately distinguish the origin of each Coptis chinensis from the high recognition rate of internal and external features Between sets of a certain degree of complementarity, a comprehensive identification of different origin Coptis Pieces of the differences. Conclusion The recognition model based on support vector machine is established to realize the distinction of origin and provide the research ideas and basis for the division of the origin of Coptis.