基于三维荧光光谱技术对掺假蜂蜜无损鉴别研究

提出了一种应用三维荧光谱技术结合化学计量学方法快速无损鉴别蜂蜜中大米糖浆掺假的新方法。利用特征参量法和主成分分析法对三维荧光光谱信息量进行压缩提取，并结合线性判别分析法(LDA)和误差反向传播神经网络法(BP-ANN)对蜂蜜掺假进行分析。结果显示，在掺假蜂蜜判别试验中，采用4个主成分时，模型对预测集样本的识别率最佳，LDA模型识别率为94.44%，BP-ANN模型识别率为100%，说明非线性的BP-ANN模型更适合蜂蜜掺假识别。研究表明，三维荧光光谱结合BP-ANN判别模型可以快速、 无损、 准确地鉴别蜂蜜中大米糖浆掺假。

Identification of Adulterated Honey Based on Three-Dimensional Fluorescence Spectra Technology

A new method was put forward to identify rice syrup adulteration in honey using three-dimensional fluorescence spectra technology combined with chemometric methods. The information of the three-dimensional fluorescence spectra was compressed and extracted using characteristic parameters method and the principal component analysis. Then, it can analyze honey adulteration combined with linear discriminant analysis (LDA) and the error back propagation neural network (BP-ANN) methods. The results showed that the prediction recognition rate of the LDA model was 94.44%, but that of the BP-ANN model can reach 100% when the number of principal components were four in the recognition experiment of adulterated honey, indicating that the BP-ANN model was more suitable for honey adulteration identification. This work demonstrated that the three-dimensional fluorescence spectroscopy combined with the BP-ANN model can be rapid, nondestructive, and accurate in identifying adulterated honey.