(1.山西农业大学农业工程学院 山西太谷 030801;2.山西农业大学食品科学与工程学院 山西太谷 030801)
(1.College of Agricultural Engineering， Shanxi Agricultural University， Taigu 030801， Shanxi ;2.College of Food Science And Engineering， Shanxi Agricultural University， Taigu 030801， Shanxi)
采用二维相关近红外光谱技术与判别分析(DA)方法相结合鉴别不同醋龄的老陈醋样品。选取5个醋龄共100个老陈醋样品，在室温下采集所有样品的一维近红外光谱。以醋龄作为外扰因素，构建各样品在1 001~1 830 nm波段范围的二维相关同步谱和异步谱，分析其二维相关谱图的光谱特征。基于二维相关同步谱的对角线数据(即自相关谱)，建立不同醋龄老陈醋样品的判别分析模型，模型校正集的识别准确率为97.3%，预测集识别准确率为96%。研究结果表明，所提出的方法可提供随醋龄变化的更多光谱特征信息，为老陈醋醋龄快速鉴别提供一种精度高的分类模型。
Two-dimensional correlation NIR spectroscopy was combined with discriminant analysis (DA) method for the identification of aged vinegar samples of different vinegar ages. A total of 100 aged vinegar samples of five different vinegar ages were selected， and the 1D NIR spectra of all samples were collected at room temperature. The two-dimensional correlation synchronous spectra and asynchronous spectra of each sample in the wavelength range of 1 001-1 830 nm were constructed with the age of vinegar as the external disturbance factor， and the spectral characteristics of the two-dimensional correlation spectra of the samples were analyzed. Based on the diagonal data (autocorrelation spectrum) of the two-dimensional correlation synchronous spectra， a discriminant analysis model was established for the old aged vinegar samples with different vinegar ages， and the identification accuracy of the model calibration set was 97.3% and that of the prediction set was 96%. The results showed that the proposed method could provide more spectral feature information with vinegar age， and provide a classification model with higher accuracy for the rapid identification of aged vinegar.