基于光谱特征提取的橡胶手套无损检验研究
Research on non-destructive testing of rubber gloves based on spectral feature extraction
田陆川 1陈浩然 1姜红 2吕航 1孙家政 1段斌 3刘峰3
作者信息
- 1. 中国人民公安大学侦查学院,北京 100038
- 2. 万子健检测技术(北京)有限公司司法鉴定中心,北京 100141
- 3. 南京简智仪器设备有限公司,江苏南京 210049
- 折叠
摘要
为建立一种快速、无损检验橡胶手套的分析,利用差分拉曼光谱仪在激发波长 785 nm,采集时间为 40 s,激光功率为 220 mW条件下,对收集到的 40个不同厂家、不同品牌、不同用途的橡胶手套样本进行检测.根据样本的主要成分不同可以将 40个样本分为四大类,可根据填料的不同对同一类样本进一步分组.结合系统聚类、Pearson相关系数法对实验数据进行处理,可根据其相关性将样本分为 14组,分组结果的DP区分能力达到 89.74%,组内样品间的相关性均大于85%,该方法可有效地对样品进行区分.
Abstract
In order to establish a rapid and non-destructive analysis of rubber gloves,40 rubber glove samples from different manufacturers,brands and uses were tested by differential Raman spectrometer under the conditions of excitation wavelength 785 nm,acquisition time 40 s and laser power 220 mW.According to the main components of the samples,40 samples can be divided into four categories,and the same type of samples can be further grouped according to different fillers.Combined with systematic clustering and Pearson correlation coefficient method to process the experimental data,the samples can be divided into 14 groups according to their correlation.The DP discrimination ability of grouping results is 89.74%,and the correlation between samples in the group is greater than 85%.This method can effectively distinguish samples.
关键词
差分拉曼光谱/橡胶手套/系统聚类/Pearson相关系数/DP值Key words
differential Raman spectroscopy/rubber glove/systematic clustering/Pearson correlation coefficient/DP discrimination引用本文复制引用
基金项目
中国人民公安大学2021年度基科费重点项目(2021JKF212)
南京简智仪器设备有限公司技术合作项目(20191218)
出版年
2024
NETL
NSTL
万方数据