首页 | 本学科首页   官方微博 | 高级检索  
     检索      

近红外光谱法结合支持向量机测定天然牛黄粉中人工牛黄的掺入量
引用本文:马群,郝贵奇,乔延江,张卓勇,张孝芳.近红外光谱法结合支持向量机测定天然牛黄粉中人工牛黄的掺入量[J].光谱学与光谱分析,2006,26(10):1842-1845.
作者姓名:马群  郝贵奇  乔延江  张卓勇  张孝芳
作者单位:1. 北京中医药大学中药学院, 北京 100102
2. 北京同仁堂股份有限公司科学研究所, 北京 100011
3. 首都师范大学化学系, 北京 100037
摘    要:提出了应用近红外漫反射光谱技术结合支持向量机测定天然牛黄粉中人工牛黄的含量的方法。以傅里叶变换近红外光谱仪(4 000 ~10 000 cm-1)为试验仪器,以含有不等量人工牛黄的天然牛黄粉(天然牛黄的质量分数范围为 0%~100%)作为校正样品,对光谱数据进行平滑、求导和小波压缩,结合支持向量机,建立了测定天然牛黄粉中人工牛黄含量的模型。试验结果为: 预测相对误差的平方和可达 0.001 35。研究表明:近红外漫反射光谱法结合支持向量机可以测定天然牛黄粉中人工牛黄的掺入量,结果可靠, 可用于天然牛黄粉的质量控制。

关 键 词:近红外漫反射  支持向量机  天然牛黄  人工牛黄  
文章编号:1000-0593(2006)10-1842-04
收稿时间:2006-01-12
修稿时间:2006-07-13

Determination of the Artificial Bezoar Powder in Bezoar Powder by Near-Infrared Spectrometry and Support Vector Machine
MA Qun,HAO Gui-qi,QIAO Yan-jiang,ZHANG Zhuo-yong,ZHANG Xiao-fang.Determination of the Artificial Bezoar Powder in Bezoar Powder by Near-Infrared Spectrometry and Support Vector Machine[J].Spectroscopy and Spectral Analysis,2006,26(10):1842-1845.
Authors:MA Qun  HAO Gui-qi  QIAO Yan-jiang  ZHANG Zhuo-yong  ZHANG Xiao-fang
Institution:1. Beijing University of Chinese Medicinal and Pharmaceutical Sciences, Beijing 100102, China2. Research Institute, Beijing Tongrentang Co., Ltd, Beijing 100011, China3. Department of Chemistry, Capital Normal University, Beijing 100037, China
Abstract:A method for determining the artificial bezoar powder in bezoar powder using near-infrared(NIR) diffuse reflectance spectrometry was proposed in the present paper.The method was based on support vector machine(SVM).The calibration set was set up by adding unequal artificial bezoar powder to the bezoar powder(content range: 0%-100%) and collecting the NIR spectrum of the samples in the wave number range of 4 000-10 000 cm~(-1).The processing algorithm was wavelet transform with first and second derivatives.A mathematical model with support vector machine was established.The model was checked with leave one method.The sum of the square of the relative prediction error was 0.001 35.This method is reliable and can be used to control the quality of bezoar powder.
Keywords:Near-infrared diffuse reflectance  Support Vector Machine  Bezoar powder  Artificial bezoar powder
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《光谱学与光谱分析》浏览原始摘要信息
点击此处可从《光谱学与光谱分析》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号