论文部分内容阅读
中药材的成分非常复杂,对中药材的质量分析一直是中药制药产业当中的重要一环,而且也是生产体系当中的重点、难点。以往传统的质量分析方法或是由于不可避免的主观因素占主导地位而可靠性较差,或是由于技术局限而破坏了中药材的整体性,因而对中药材进行客观的整体分析就显得极为重要。与此同时,傅里叶变换红外光谱法因其快速、准确、无损检测等特点正受到越来越多的关注,在很多行业领域中也已得到广泛的应用。本文创新地运用红外光谱技术结合模式识别技术,分别对不同产地和不同等级的丹参药材进行了建模分析,并在此基础上评价了3种常用的聚类算法——支持向量机(SVM)、自适应提升算法(AdaBoost)以及线性判别函数(LDF)——在丹参药材模型建立以及聚类分析中的应用效果。结果表明,LDF对不同产地丹参药材聚类模型的识别率和拒绝率最高;而等级分类建模方面则以SVM的识别率和拒绝率最高,且均可达到97%以上。由此可以得出结论,与传统的质量分析方法相比,红外光谱与模式识别相结合的新分析技术是解决中药整体分析的有效方法之一。
The composition of Chinese herbal medicines is very complex. The quality analysis of Chinese herbal medicines has always been an important part of traditional Chinese medicine and pharmaceutical industry, and it is also the key and difficult point in the production system. In the past, the traditional method of quality analysis was either based on the unreliability of the dominant subjective factors and the poor reliability or because of the technical limitations of the integrity of Chinese herbal medicines, and therefore an objective and overall analysis of Chinese herbal medicines is of paramount importance . At the same time, Fourier transform infrared spectroscopy is receiving more and more attention because of its fast, accurate and non-destructive testing. It has also been widely used in many fields of industry. This paper innovatively uses infrared spectroscopy combined with pattern recognition technology to model and analyze Salvia miltiorrhiza medicinal materials in different producing areas and different grades. Based on this, three commonly used clustering algorithms - support vector machine (SVM) , AdaBoost and Linear Discriminant Function (LDF) - in the establishment of Salvia miltiorrhiza medicinal herbs model and cluster analysis. The results showed that LDF had the highest recognition rate and rejection rate for Salvia miltiorrhiza medicinal cluster clustering models in different origins, while the SVR classification rate and rejection rate were the highest in LDF classification, both of which reached more than 97%. It can be concluded that, compared with the traditional methods of mass analysis, the new analytical technique combining infrared spectroscopy and pattern recognition is one of the effective methods to solve the overall analysis of traditional Chinese medicine.