一种基于最小二乘支持向量机算法的近红外光谱判别分析方法

Discriminating the Genuineness of Chinese Medicines with Least Squares Support Vector Machines

将最小二乘支持向量机(LSSVM)用于近红外(NIR)光谱分析,建立一种新型的NIR光谱快速鉴别方法。以丹参药材道地性鉴别为例,对其NIR漫反射光谱进行主成分分析后,运用LSSVM法建立NIR光谱非线性分类模型,对丹参药材道地性进行快速鉴别。将本方法与经典SVM和BP神经网络法相比较,结果表明,本法判别准确率高,计算时间少,可推广应用于中药等天然产物质量快速鉴别。

A method for the rapid identification of the genuineness of Chinese medicines based on near infrared （NIR） spectroscopy and least square support vector machines （LSSVM） was proposed. In this study, NIR spectra of the powdered Danshen （Radix Salviae Miltiorrhizae） were collected, and the nonlinear classifier based on LSSVM algorithm was developed to discriminate thegenuineness of these herbs. The result obtained by the proposed method was compared with that from the traditional support vector machine （SVM） and back propagation-artificial neural network （BP-ANN） methods. It was shown that the generalizationperformance of the classifier based on LSSVM was much better than that of BP-ANN, while thecomputation time of LSSVM is much shorter than that of the traditional SVM. The method proposed can be applied to the rapid and accurate identification of the quality of the naturalproducts.