|
二维码(扫一下试试看!) |
基于主成分分析法对一次性纸杯的拉曼光谱检验研究 |
Detection of Disposable Paper Cup by Raman Spectroscopy and PCA |
投稿时间:2020-06-17 |
DOI:10.11980/j.issn.0254-508X.2020.09.007 |
关键词: 一次性纸杯 拉曼光谱 主成分分析法 聚类分析 |
Key Words:disposable paper cup Raman spectroscopy principal component analysis cluster analysis |
基金项目:中国人民公安大学2019年度基科费重点项目(2019JKF222);国家重点研发计划项目(2017YFC0822004)。 |
|
摘要点击次数: 3915 |
全文下载次数: 2120 |
摘要:为建立一种检验犯罪现场遗留一次性纸杯物证的科学有效的方法,利用拉曼光谱结合化学计量学对收集到的27个不同品牌、不同用途的一次性纸杯样本的拉曼特征峰峰位和峰强度的对比,对一次性纸杯样本进行区分。为了使分类结果科学准确,利用化学计量学的方法对光谱数据进行分析处理。为了减少聚类分析所用的变量个数,利用主成分分析法对拉曼光谱数据进行降维处理,指定提取3个因子,减少变量个数,保留大部分信息,其累计贡献率达99.09%。利用层次聚类和K-Means快速聚类将27个样本分为8类,并利用Fisher判别分析法验证了分类结果的科学合理性,从而对未知样本的类别判断提供依据。 |
Abstract:In order to establish a scientific and effective method to test the physical evidence of disposable paper cup left at the scene of crime, Raman spectroscopy and chemometrics were used to test 27 samples of disposable paper cups of different brands and different uses. The Raman characteristic peaks and peak intensities of different disposable paper cup samples were not the same, so the purpose of distinguishing could be achieved. Principal component analysis (PCA) was used to reduce the dimension of the full wave band Raman spectrum data. Three factors were designated to extract, which accounted for 99.09% of the original variables. The number of variables were reduced and most of the information was protected. Based on this, using hierarchical clustering and K-means fast clustering, 27 samples were divided into 8 categories, and Fisher discriminant analysis was used to verify the scientific rationality of the classification results, so as to provide the basis for the classification of unknown samples. |
查看全文 查看/发表评论 下载PDF阅读器 |