基于主成分分析-Fisher判别分析的食品类塑料瓶物证差分拉曼光谱分类

食品类塑料瓶物证携带许多潜在证据信息，目前针对此类物证的检验研究尚处于探索阶段。利用差分拉曼光谱对46个食品类塑料瓶样品进行检验，依据样品材质及光谱特征峰可将样品分为三类。利用主成分分析（Principal component analysis, PCA）-Fisher判别分析，绘制主成分得分图，构建判别函数，建立分类模型。结果表明，食品类塑料瓶样品具有明显的聚类关系，原始分类与交叉验证分类准确率达到100 %。差分拉曼光谱结合PCA-Fisher判别分析，检验鉴别食品类塑料瓶物证具有一定的科学性。

DifferentialRamanSpectralInspectionofFoodGradePlasticBottlesBasedonPrincipalComponentAnalysisandFisherDiscriminantAnalysis

The material evidence of plastic food grade bottles carries many potential evidentiary information, and the current research on the inspection of such material evidence is still in the exploratory stage. Using differential Raman spectroscopy, 46 food grade plastic bottle samples were examined. Samples can be divided into three categories based on their material and spectral characteristic peaks. Using principal component analysis - Fisher discriminant analysis, draw a principal component score map, construct a discriminant function, and established a classification model. The results showed that the food grade plastic bottle samples had a significant clustering relationship, and the accuracy rate of the original classification and cross validation classification reached 100%. Differential Raman spectroscopy combined with PCA-Fisher discriminant analysis has certain scientific significance in testing and identifying the material evidence of food grade plastic bottles.