基于化学计量学的一次性塑料餐盒红外光谱分析

Infrared Spectroscopic Analysis of Disposable Plastic Food Box Based on Chemometrics

利用红外光谱法对35个一次性塑料餐盒样本进行了谱图的无损采集。经过前期Savitzky-Golay平滑、多元散射校正等数据预处理后,利用样本之间的余弦相似度进行初步分类。结合主成分分析法提取样本主成分,根据样本的主成分得分进行分类,利用相关性分析对于不同分类结果做出解释和取舍。最后利用Fisher判别分析对样本分类情况进行检验,通过分析各类别重心分布情况和类间距离考察分类效果,判别分析中留一交叉验证正确率为97.1%,证明判别模型分类效果良好,可为未知样本模式识别提供模型基础。提供了一种普遍性较强的模型分类方法,对于物证的模式识别有一定参考意义。

Non-destructive inspection and accurate classification of physical evidence has always been the focus in the field of physical and chemical evidence inspection.By analyzing the physical evidence of disposable plastic food boxes left at the scene,clues and directions can be provided for public security departments to detect cases.In the experiment,Infrared spectra of 35 disposable plastic food box samples were collected nondestructively.After pre-processing of data such as Savitzky-Golay smoothing and multiplicative scatter correction,the Cosine Similarity among samples was used for preliminary classification.Principal Component Analysis(PCA)was used to extract the principal component of samples,and the classification is based on the principal component scores of the samples.Correlation analysis was used to explain and choose different classification results.Finally,Fisher discriminant analysis was used to test the classification of samples.By analyzing the distribution of the center of gravity and the distance between the classes,the classification effect was analyzed.The accuracy of Leave-One-Out-Cross-Validation was 97.1%in the discriminant analysis,which proved that the classification effect of the discriminant model was effective and could provide a model foundation for pattern recognition of unknown samples.The experiment provides a general model classification method,which has certain reference significance for pattern recognition of physical evidence.