• 首页期刊简介编委会刊物订阅专栏专刊电子刊学术动态联系我们English
引用本文:孙梅,王睿,陈兴海,黄宇,刘业林.高光谱在冬虫夏草含量及真伪鉴别中的应用[J].中国现代应用药学,2017,34(8):1114-1117.
SUN Mei,WANG Rui,CHEN Xinghai,HUANG Yu,LIU Yelin.Application of Hyperspectral Imaging Technique for Cordyceps Powder Content and True/False Identification[J].Chin J Mod Appl Pharm(中国现代应用药学),2017,34(8):1114-1117.
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 2191次   下载 1730 本文二维码信息
码上扫一扫!
分享到: 微信 更多
高光谱在冬虫夏草含量及真伪鉴别中的应用
孙梅1, 王睿1, 陈兴海2, 黄宇2, 刘业林2
1.北京工商大学计算机与信息工程学院, 北京 100048;2.四川双利合谱科技有限公司, 成都 610016
摘要:
目的 以冬虫夏草粉末为研究对象,利用高光谱成像技术建立对冬虫夏草粉末的真假鉴别及含量判断的无损检测模型。方法 将真伪样品粉末分别按9︰1、8︰2、7︰3、6︰4、5︰5等不同比例进行混合,通过光谱范围为940~2 500 nm的高光谱成像仪分别获取真伪样品、不同比例真伪混合样品的高光谱影像数据,经过主成分分析(PCA)初步判别真伪样品的差异性,再结合偏最小二乘法(PLS)对真伪粉末样品的含量进行分析判断。结果 不同比例真伪混合物中冬虫夏草粉末的识别度为97.0%~98.78%,伪冬虫夏草粉末的识别度为83.10%~99.3%。结论 基于高光成像技术可以实现对冬虫夏草粉的真假辨别,并可准确地判别出冬虫夏草粉末的有效含量。
关键词:  高光谱成像技术  冬虫夏草粉末  真假鉴别  含量判断
DOI:10.13748/j.cnki.issn1007-7693.2017.08.009
分类号:R917.102
基金项目:北京工商大学两科基金培育项目(RJJ2016-21)
Application of Hyperspectral Imaging Technique for Cordyceps Powder Content and True/False Identification
SUN Mei1, WANG Rui1, CHEN Xinghai2, HUANG Yu2, LIU Yelin2
1.School of Computer and Information Engineering, Beijing Technology and Business University, Beijing 100048, China;2.Sichuan Dualix Spectral Image Technology Co., Ltd., Chengdu 610016, China
Abstract:
OBJECTIVE To identify true or false of cordyceps powder and judge it's content use the hyperspectral imaging technology. METHODS The true and false samples powder were mixed at different ratios of 9:1, 8:2, 7:3, 6:4, 5:5, respectively. The hyperspectral image data of samples and different proportions of true and false mixed samples were obtained by the spectral range of 940-2 500 nm hyperspectral imager. Through principal component analysis (PCA) to determine the difference between true and false samples, and combined with partial least squares (PLS) method to analyze the content of true and false powder samples. RESULTS In different proportions of true and false mixtures, cordyceps powder identification degree was 97.0%-98.78%, and false cordyceps powder identification degree was 83.10%-99.3%. CONCLUSION The research results show it is convenient to identify the true and false of cordyceps powder by the hyperspectral imaging technology, and it is suitable for quality evaluation of traditional cordyceps powder.
Key words:  hyperspectral imaging  cordyceps powder  true and false identification  content measurement
扫一扫关注本刊微信