摘要
针对山西醋业市场存在的问题,采用可见-近红外光谱技术,结合化学计量学方法,开展山西老陈醋产地的判别分类工作。对3个不同产地的240个山西老陈醋样品采集光谱数据,分析不同预处理方法(未处理、平滑、微分)对陈醋产地判别结果的影响;结合判别分析法建立陈醋产地定性判别模型。试验结果表明,可见-近红外原始光谱结合判别分析法的结果最优,校正集正确分类的比例达98.46%,预测集达93.33%;微分光谱分析结果最差。
In view of existing problem at Shanxi vinegar market,the discriminant analysis of Shanxi aged vinegars producing area was carried out by vis/NIR spectroscopy combined with chemometrics method.Spectral data of 240 aged vinegars samgles from 3 producing area were collected.The effect of different pretreatment methods(no treatment,smooth treatment,second derivative) on discriminatory result of aged vinegars producing area was analysed,and the qualitative discrimintion model of aged vinegar producing area was also established.The results showed that the optimum analytical result was obtained by original spectrum of Vis/NIR spectroscopy combined with discriminatory analysis.Its correct classification percentage was reached to 98.46% for calibration set 99.33% for validation set.The analytical results of derivative spectrum was the worst.
出处
《中国食品学报》
EI
CAS
CSCD
北大核心
2010年第4期267-271,共5页
Journal of Chinese Institute Of Food Science and Technology
基金
山西省科技攻关项目(20100311066-5)
关键词
近红外透射光谱
判别分析
陈醋
产地
定性分析
Vis/NIR transmission spectroscopy
aged vinegar
discriminant analysis
geographical origin
qualitative analysis