摘要
为了实现对掺假芝麻油的快速鉴别,将豆油、花生油、棉籽油和菜籽油分别与同一种纯正芝麻油按体积分数0.5%～100%的比例混合,在3 200～10 000cm-1光谱范围内采集了掺假芝麻油样品的近红外吸收光谱。通过特征谱区的选择、光谱预处理方法的优化,采用聚类分析和主成分分析法(PCA)分别建立了芝麻油的鉴别模型。结果表明:4种掺假油品有不同的较优光谱处理范围;两种模式识别方法对于掺假量5%～100%的芝麻油真伪识别率达到100%;而掺假量在5%以下时,两种方法都失去鉴别能力,说明近红外光谱分析技术在检测掺假芝麻油时的最低掺假下限为5%。综上,近红外光谱结合模式识别技术在掺假量大于5%时,可快速、准确地鉴别真伪芝麻油。
In order to realize the fast identification of adulteration sesame oil, this study will soybean oil, peanut oil, cottonseed oil and colza oil respectively with the same kind of pure sesame oil by volume fraction of 0.5% to 100% of the mixing ratio and collected the near infrared absorption spectrum of adulteration sesame oil samples in 3 200 ~ 10 000 cm-I spectral range. Through the spectrum characteristics choose and spectral pretreatment methods of optimization,This study adopting clustering analysis and principal component analysis （PCA）respectively establish sesame oil identified model. The result showed that four adulteration oil have different is optimal spectrum processing range; The identifying rate of 5% ~ 100% adulterated samples tested by two models on the basis of pattern recognition was 100% ; The amount of adulteration below 5% ,two kinds of methods are losing ability to identify that of near in- frared spectral analysis technology in the detection of adulteration sesame minimum lower quality for 5%. In conclu- sion,the near infrared spectrum combined with pattern recognition technology in spoofing quantity is more than 5%, can be used to quickly and accurately identify true bogus sesame.
出处
《中国粮油学报》
EI
CAS
CSCD
北大核心
2012年第12期116-121,共6页
Journal of the Chinese Cereals and Oils Association
关键词
近红外光谱
芝麻油
掺假
聚类分析
主成分分析
near - infrared spectroscopy, sesam oil, adulteration, clustering analysis, principal component analysis （PCA）
