光谱学与光谱分析
光譜學與光譜分析
광보학여광보분석
SPECTROSCOPY AND SPECTRAL ANALYSIS
2013年
11期
3032-3035
,共4页
苗静%曹玉珍%杨仁杰%刘蓉%孙惠丽%徐可欣
苗靜%曹玉珍%楊仁傑%劉蓉%孫惠麗%徐可訢
묘정%조옥진%양인걸%류용%손혜려%서가흔
二维相关近红外谱%参数化%掺杂牛奶%BP神经网络%尿素%三聚氰胺
二維相關近紅外譜%參數化%摻雜牛奶%BP神經網絡%尿素%三聚氰胺
이유상관근홍외보%삼수화%참잡우내%BP신경망락%뇨소%삼취청알
Two-dimensional correlation near-infrared spectra%Parameterization%Adulterated milk%BP neural network%Urea%M elamine
将二维相关近红外谱参数化方法与BP神经网络结合,建立掺杂牛奶与纯牛奶的判别模型。分别配制含有尿素牛奶(1~20g·L -1)和三聚氰胺牛奶(0.01~3g·L -1)样品各40个。研究了纯牛奶、掺杂牛奶的二维相关近红外谱特性,在此基础上,分别提取了各样品二维相关同步谱的5个特征参数。将这5个特征参数作为BP神经网络的输入,分别建立掺杂尿素、掺杂三聚氰胺、两种掺杂牛奶与纯牛奶的判别模型,采用这些模型对未知样品进行预测,其预测正确率分别为95%,100%和96.7%。研究结果表明:该方法有效地提取了牛奶中掺杂目标物的特征光谱信息,同时又减少了BP神经网络输入变量的维数,实现了掺杂牛奶与纯牛奶的鉴别。
將二維相關近紅外譜參數化方法與BP神經網絡結閤,建立摻雜牛奶與純牛奶的判彆模型。分彆配製含有尿素牛奶(1~20g·L -1)和三聚氰胺牛奶(0.01~3g·L -1)樣品各40箇。研究瞭純牛奶、摻雜牛奶的二維相關近紅外譜特性,在此基礎上,分彆提取瞭各樣品二維相關同步譜的5箇特徵參數。將這5箇特徵參數作為BP神經網絡的輸入,分彆建立摻雜尿素、摻雜三聚氰胺、兩種摻雜牛奶與純牛奶的判彆模型,採用這些模型對未知樣品進行預測,其預測正確率分彆為95%,100%和96.7%。研究結果錶明:該方法有效地提取瞭牛奶中摻雜目標物的特徵光譜信息,同時又減少瞭BP神經網絡輸入變量的維數,實現瞭摻雜牛奶與純牛奶的鑒彆。
장이유상관근홍외보삼수화방법여BP신경망락결합,건립참잡우내여순우내적판별모형。분별배제함유뇨소우내(1~20g·L -1)화삼취청알우내(0.01~3g·L -1)양품각40개。연구료순우내、참잡우내적이유상관근홍외보특성,재차기출상,분별제취료각양품이유상관동보보적5개특정삼수。장저5개특정삼수작위BP신경망락적수입,분별건립참잡뇨소、참잡삼취청알、량충참잡우내여순우내적판별모형,채용저사모형대미지양품진행예측,기예측정학솔분별위95%,100%화96.7%。연구결과표명:해방법유효지제취료우내중참잡목표물적특정광보신식,동시우감소료BP신경망락수입변량적유수,실현료참잡우내여순우내적감별。
Discriminant models of adulterated milk and pure milk were established using BP neural network combined with two-dimensional (2D) correlation near-infrared spectra parameterization .Forty pure milk samples ,40 adulterated milk samples with urea (1~20 g · L -1 ) and 40 adulterated milk samples with melamine (0.01~3 g · L -1 ) were prepared respectively .Based on the characteristics of 2D correlation near-infrared spectra of pure milk and adulterated milk ,5 apparent statistic parameters were calculated based on the parameterization theory .Using 5 characteristic parameters ,discriminant models of urea adulterated milk , melamine adulterated milk and two types of adulterated milk were built by BP neural network .The prediction rate of unknown samples were 95% ,100% and 96.7% ,respectively .The results show that this method can extract effectively feature informa-tion of adulterant ,reduce the input dimensions of BP neural network ,and better realize qualitative analysis of adulterant in milk .