摘要
目的淫羊藿作为一种药食两用植物,在祛湿补肾方面有显著疗效。应用近红外光谱,以甘肃、陕西和辽宁三个不同产区的淫羊藿为研究对象进行产地鉴别。方法采用主成分分析(principal component analysis,PCA)、前馈人工神经网络(back feed forward-artificial neural network,BP-ANN)和支持向量机(support vector machine,SVM)进行定性判别分析;其中,在支持向量机分类模型中,研究了三种参数寻优方法包括网格全局搜索(grid search)、遗传算法(genetic algorithm,GA)及粒子群算法(particle swarm optimization,PSO)对模型性能的影响。结果 PCA得分图产地间有部分重叠,较难区分;前馈人工神经网络和支持向量机定性识别方法都能完全准确地鉴别产地。结论该研究表明近红外光谱技术结合化学计量学可作为一种快速可靠的方法用于淫羊藿产地的鉴别,并为市场规范提供一种新思路。
Objective Herba Epimedii is a kind of medical and edible plants and it exhibits a satisfactory efficacy on tonifying the kidney and dispelling dampness. The Herba Epimedii samples from Gansu,Shanxi and Liaoning were collected to identify their geographical origins with the application of near infrared spectroscopy. Methods Principal component analysis（ PCA）,back feed forward-artificial neural network（ BP-ANN） and support vector machine（ SVM） were used to qualitative discriminant analysis. In addition,the efficacy of the parameter optimization method such as grid search methods,genetic algorithm（ GA） and particle swarm optimization（ PSO） on the performance of the SVM model was investigated. Results The results showed that the scores plot of their principal components somewhat overlapped and the origins of the Herba Epimedii were difficult to be identified,while the classification accuracy of BP-ANN and SVM could be up to 100%. Conclusion The study revealed that the near infrared spectroscopy combined with chemometrics can be used as a fast and reliable method for the origin identification of the Herba Epimedii.
出处
《时珍国医国药》
CAS
CSCD
北大核心
2017年第8期1902-1905,共4页
Lishizhen Medicine and Materia Medica Research
基金
国家科技重大专项(No.2014ZX09201021-010)
关键词
近红外光谱
淫羊藿
产地鉴别
支持向量机
参数优化
Near infrared spectroscopy
Herba Epimedii
Origin identification
Support vector machine
Parameter optimization
