CAJ | 학술논문

采用近红外光谱分析技术结合化学计量学方法研究对不同品牌的豆浆粉以及假冒的豆浆粉鉴别的可行性。采集不同品牌豆浆粉以及假冒豆浆粉在12500～4000 cm -1范围内光谱，并进行不同的预处理。采用偏最小二乘-判别分析（partial least squares-discriminant analysis ，PLS-DA ）对不同预处理的光谱进行建模比较，去趋势算法（De-trending）预处理光谱与多元散射校正（multiplicative scatter correction ，MSC）结合De-trending（MSC＋De-trending）预处理光谱的 PLS-DA模型预测集判别正确率最高，均为100％。采用 x-loading weights方法分别基于De-trending和MSC-De-trending预处理光谱选择了6个和7个特征波数，并以特征波数分别建立了线性判别分析（linear discriminant analysis ，LDA ）和误差反向传播神经网络（back-propagation neural network ，BPNN）的判别分析模型。结果表明，以所选出的不同的特征波数建立的BPNN判别分析模型取得了最佳的判别效果，建模集和预测集的判别正确率均为100％。采用近红外光谱分析技术可以准确的判别豆浆粉品牌以及假冒豆浆粉产品。
채용근홍외광보분석기술결합화학계량학방법연구대불동품패적두장분이급가모적두장분감별적가행성。채집불동품패두장분이급가모두장분재12500～4000 cm -1범위내광보，병진행불동적예처리。채용편최소이승-판별분석（partial least squares-discriminant analysis ，PLS-DA ）대불동예처리적광보진행건모비교，거추세산법（De-trending）예처리광보여다원산사교정（multiplicative scatter correction ，MSC）결합De-trending（MSC＋De-trending）예처리광보적 PLS-DA모형예측집판별정학솔최고，균위100％。채용 x-loading weights방법분별기우De-trending화MSC-De-trending예처리광보선택료6개화7개특정파수，병이특정파수분별건립료선성판별분석（linear discriminant analysis ，LDA ）화오차반향전파신경망락（back-propagation neural network ，BPNN）적판별분석모형。결과표명，이소선출적불동적특정파수건립적BPNN판별분석모형취득료최가적판별효과，건모집화예측집적판별정학솔균위100％。채용근홍외광보분석기술가이준학적판별두장분품패이급가모두장분산품。
Near-infrared spectroscopy combined with chemometrics was used to investigate the feasibility of identifying different brands of soymilk powder and the counterfeit soymilk powder products .For this purpose ,partial least squares-discriminant anal-ysis (PLS-DA) ,linear discriminant analysis (LDA) and back-propagation neural network (BPNN) were employed as pattern recognition methods to class ify soymilk powder samples .The performances of different pretreatments of raw spectra were also compared by PLS-DA .PLS-DA models based on De-trending and multiplicative scatter correction (MSC)combined with De-tren-ding(MSC+ De-trending) spectra obtained best results with 100% prediction accuracy ,respectively .Six and seven optimal wavenumbers selected by x-loading weights of the best two PLS-DA models were used to build LDA and BPNN models .Results showed that BPNN performed best and correctly classified 100% of the soymilk powder samples for both the calibration and the prediction set .The overall results indicated that NIR spectroscopy could accurately identify branded and counterfeit soymilk powder products .