光谱技术结合化学计量学识别苹果品种

Identification of Apple Varieties Based on Spectroscopy Technology Combined With Chemometrics

以“冰糖心”“嘎啦”和“山东富士”苹果为试材,采用紫外/可见光谱技术结合化学计量学,分别建立了判别3种苹果品种的K最近邻(KNN)识别模型和偏最小二乘判别分析(PLS-DA)识别模型,分析了不同的光谱预处理方法(二阶微分(SD)、标准正态变换(SNV)和多元散射校正(MSC))对各模型识别效果的影响,并采用主成分分析方法对预处理后的光谱数据进行降维,以提取能反映苹果品种的特征光谱。结果表明:采用主成分分析法选择了累计贡献率超过99%的前6个主成分作为样本集特征光谱数据,很好地实现了光谱数据的降维;二阶微分预处理方法对光谱的预处理效果最好。综上所述,建立的识别模型均能基本满足实际要求,且SD+KNN模型的建模效果最好,MSC+KNN模型的预测性能最好,SD+PLS-DA模型的总识别效果最好。

The recognition models of three different apple varieties(‘Candy’‘Gala’ and ‘Fuji’ apple) were established based on UV/VIS spectroscopy technology combined with K nearest neighbor(KNN) and partial least-square discriminant analysis(PLS-DA),respectively.Then the influence of the different spectrum pretreatment methods(second derivation(SD),standard normal variation(SNV) and multi-scatter calibration(MSC)) on the recognition models was analyzed.Finally,the characteristic spectrum of apples were extracted by the principal component analysis.The results showed that the first six principal components with the cumulative contribution rate above 99% were selected as the characteristic spectral data in the sample set by the principal component analysis,and the dimensionality reduction of the spectral data was well achieved.The effect of second derivation preprocessing for the spectrum data was the best.In conclusion,the recognition models all had an acceptable accuracy,especially SD+KNN model had optimal recognition performance for modeling set and MSC+KNN model had optimal recognition performance for prediction set.And SD+PLS-DA model had optimal total recognition performance.