光谱学与光谱分析
光譜學與光譜分析
광보학여광보분석
SPECTROSCOPY AND SPECTRAL ANALYSIS
2014年
5期
1259-1263
,共5页
刘星%毛丹卓%王正武%杨永健
劉星%毛丹卓%王正武%楊永健
류성%모단탁%왕정무%양영건
薏仁%近红外光谱%支持向量机%学习向量量化神经网络%定性判别
薏仁%近紅外光譜%支持嚮量機%學習嚮量量化神經網絡%定性判彆
의인%근홍외광보%지지향량궤%학습향량양화신경망락%정성판별
Coix seed%Near infrared spectroscopy%Support vector machine%Learning vector quantization neural network%Qualitative discriminant
薏仁是一种药食两用资源,对其品质快速鉴别的需求也越来越多,近红外光谱技术(near inf rared spectroscopy ,NIRS)作为一种快速、无损且环保的方法正适合这一需求。以不同产地和品种薏仁的近红外光谱为基础,结合化学计量学方法对薏仁种类进行鉴别。对原光谱用无监督学习算法主成分分析(principal component analysis ,PCA)和有监督学习算法学习向量量化(learning vector quantization ,LVQ)神经网络、支持向量机(support vector machine ,SVM )进行定性判别分析。由于不同地区和不同品种的薏仁营养物质组成复杂且含量相近,所选两类薏仁的特征变量很相似,因而PCA得分图重叠严重,很难区分;而LVQ神经网络和SVM都能得到满意结果,LVQ神经网络的预测正确率为90.91%,SVM 在经过惩罚参数和核函数参数优选后,分类准确率能达到100%。结果表明:近红外光谱技术结合化学计量学方法可作为一种快速、无损、可靠的方法用于薏仁种类的鉴别,并为市场规范提供技术参考。
薏仁是一種藥食兩用資源,對其品質快速鑒彆的需求也越來越多,近紅外光譜技術(near inf rared spectroscopy ,NIRS)作為一種快速、無損且環保的方法正適閤這一需求。以不同產地和品種薏仁的近紅外光譜為基礎,結閤化學計量學方法對薏仁種類進行鑒彆。對原光譜用無鑑督學習算法主成分分析(principal component analysis ,PCA)和有鑑督學習算法學習嚮量量化(learning vector quantization ,LVQ)神經網絡、支持嚮量機(support vector machine ,SVM )進行定性判彆分析。由于不同地區和不同品種的薏仁營養物質組成複雜且含量相近,所選兩類薏仁的特徵變量很相似,因而PCA得分圖重疊嚴重,很難區分;而LVQ神經網絡和SVM都能得到滿意結果,LVQ神經網絡的預測正確率為90.91%,SVM 在經過懲罰參數和覈函數參數優選後,分類準確率能達到100%。結果錶明:近紅外光譜技術結閤化學計量學方法可作為一種快速、無損、可靠的方法用于薏仁種類的鑒彆,併為市場規範提供技術參攷。
의인시일충약식량용자원,대기품질쾌속감별적수구야월래월다,근홍외광보기술(near inf rared spectroscopy ,NIRS)작위일충쾌속、무손차배보적방법정괄합저일수구。이불동산지화품충의인적근홍외광보위기출,결합화학계량학방법대의인충류진행감별。대원광보용무감독학습산법주성분분석(principal component analysis ,PCA)화유감독학습산법학습향량양화(learning vector quantization ,LVQ)신경망락、지지향량궤(support vector machine ,SVM )진행정성판별분석。유우불동지구화불동품충적의인영양물질조성복잡차함량상근,소선량류의인적특정변량흔상사,인이PCA득분도중첩엄중,흔난구분;이LVQ신경망락화SVM도능득도만의결과,LVQ신경망락적예측정학솔위90.91%,SVM 재경과징벌삼수화핵함수삼수우선후,분류준학솔능체도100%。결과표명:근홍외광보기술결합화학계량학방법가작위일충쾌속、무손、가고적방법용우의인충류적감별,병위시장규범제공기술삼고。
Unsupervised learning algorithm-principal component analysis (PCA) ,and supervised learning algorithm-learning vec-torquantization (LVQ)neuralnetworkandsupportvectormachine (SVM)wereusedtocarryoutqualitativediscriminantanaly-sis of different varieties of coix seed from different regions .Since nutrient compositions of different varieties coix seed samples from different origins were complex and the contents were similar ,characteristic variables of two kinds of coix seed were alike , the scores plot of their principal components seriously overlapped and the categories of coix seed were difficult to distinguish . While satisfactory results were obtained by LVQ neural network and SVM .The accuracy of LVQ neural network prediction is 90.91% ,while the classification accuracy of SVM ,whose penalty parameter and kernel function parameter were optimized ,can be up to 100% .The results show that NIRS combined with chemometrics can be used as a rapid ,nondestructive and reliable method to identify coix seed varieties and provide technical reference for market regulation .