基于Bayes判别的手帕纸红外光谱鉴别  

Infrared spectrum identification of handkerchief paper based on bayes discrimination

作　　者：卫辰洁 王继芬[1] 季佳华 苏东斌 马体 WEI Chen-jie;WANG Ji-fen;JI Jia-hua;SU Dong-bin;MA Ti(School of Investigation,People’s Public Security University of China,Beijing 102600,China)

机构地区：[1]中国人民公安大学侦查学院,北京102600

出　　处：《化学研究与应用》2021年第2期269-275,共7页Chemical Research and Application

基　　金：中国人民公安大学基本科研业务费重点项目(2019JKF223)资助。

摘　　要：手帕纸是犯罪现场常见的物证之一,在法庭科学领域备受关注。为了实现对市场上手帕纸的快速分类鉴别的目的,本文采用了具有无损检验特点的傅里叶红外光谱,结合主成分分析(PCA)与Bayes判别对8种品牌96个手帕纸样本建立分类模型。结果表明,分别利用PCA和Bayes判别对样本进行分类的准确率并不理想,采用Bayes判别对PCA降维后的特征向量进行建模,分类准确率可达100%。采用傅里叶红外光谱结合Bayes判别实现了对手帕纸的精准分类,为实践中手帕纸的分类鉴别提供一定的参考和借鉴。Handkerchief paper is one of the common physical evidence at the crime scene,and it has attracted much attention in the field of forensic science.In order to achieve the purpose of rapid classification and identification of handkerchief paper on the market.In this paper,Fourier infrared spectroscopy with non-destructive testing characteristics,combined with principal component analysis(PCA)and Bayes discrimination,are used to establish a classification model of 96 handkerchief paper samples from 8 brands.The results show that the accuracy rates classified by PCA and Bayes respectively are not ideal.Bayes discrimination is used to model the feature vectors based on PCA dimensionality reduction,and the classification accuracy rate can reach 100%.The use of Fourier infrared spectroscopy combined with Bayes discrimination realizes the accurate classification of handkerchief paper,which provides a reference for the classification of handkerchief paper in practice.

关 键 词：光谱学 傅里叶红外光谱 手帕纸 Bayes判别 主成分分析 

分 类 号：O657.33[理学—分析化学]