论文部分内容阅读
太赫兹时域光谱技术(THz-TDS)结合主成分分析-线性判别分析(PCA-LDA)和支持向量机(SVM)用于正品大黄样品的鉴定。在时域测量41个大黄样品的太赫兹时域透射光谱,然后将这些时域信号转换成频域的吸收系数系数。根据样本的吸收系数建立了主成分分析-线性判别分析和支持向量机的定性分类模型,并对正品和非正品大黄样本的分类模型进行了交叉验证。模型的预测能力和稳定性使用自助拉丁配分进行评价,使用50次自助拉丁配分,配分数为4。使用主成分分析-线性判别分析和支持向量机均得到了满意的结果。提出的方法证明是一种方便、无污染、准确和无需化学处理的鉴定大黄样本的方法。该文提出的步骤可以应用于其他中草药分类和生产的质量控制。
THz-TDS combined with PCA-LDA and SVM was used to identify genuine samples of rhubarb. The terahertz time-domain transmission spectra of 41 rhubarb samples were measured in the time domain, and these time domain signals were then converted into absorption coefficient coefficients in the frequency domain. According to the absorption coefficients of the samples, the principal component analysis - linear discriminant analysis and SVM classification model were established, and the classification models of genuine and non-genuine rhubarb samples were cross-validated. The predictive power and stability of the model were evaluated using a Latin American self-service allocation of 50 Latin American latin ingredients with a score of 4. Using principal component analysis - linear discriminant analysis and support vector machines have been satisfactory results. The proposed method proves to be a convenient, non-contaminated, accurate and chemical-free method for identifying rhubarb samples. The steps proposed in this paper can be applied to the quality control of other Chinese herbal medicines for classification and production.