基于迁移学习的新精神活性物质拉曼光谱识别研究  

Research on Raman spectral identification of new psychoactive substances based on transfer learning

作　　者：何洪源[1] 吕铷麟 徐琳 赵霞 魏育新[1] 师博远 HE Hongyuan;LV Rulin;XU Lin;ZHAO Xia;WEI Yuxin;SHI Boyuan(School of Investigation,People's Public Security University of China,Beijing 100038,China;Yantai City Public Security Bureau,Marine Ecological Civilization and Economy Branch,Yantai Shandong 265800,China;National Anti-Drug Laboratory Beijing Regional Center,Beijing 100164,China)

机构地区：[1]中国人民公安大学侦查学院,北京100038 [2]烟台市公安局长岛海洋生态文明经济分局,山东烟台265800 [3]国家毒品实验室北京分中心,北京100164

出　　处：《激光杂志》2023年第5期29-36,共8页Laser Journal

基　　金：上海市现场物证重点实验室开放课题(No.2021XCWZK05);中央高校基本科研业务费项目(No.2020JKGF102);国家安全学高精尖学科一般科研项目(No.2021YBKYXM023);中国人民公安大学基本科研业务费用项目(No.2021JKF103)。

摘　　要：随着精神活性物质(NPS)对社会的潜在危害越来越严重,提高新型毒品的快速检测、识别能力是当前打击毒品犯罪的重中之重。本研究采集了196份新精神活性物质缴获样本拉曼光谱,结合连续小波变换和卷积神经网络迁移学习模型,对五种检测仪器获得的拉曼光谱数据进行分析和识别。对VGG16、InceptionV3和ResNet50三种深度学习模型分类效能的比较结果显示,单仪器新精神活性物质识别能力为80.2%~100%不等,多仪器综合识别能力为88.6%。通过合理的光谱预处理方法,能够将不同仪器的NPS光谱检测数据,统一格式并进行批量的分析和识别,精准提高了拉曼光谱的数据提取和利用效率。As the potential harm of psychoactive substances(NPS)to society is becoming more and more serious,improving the rapid detection and identification of new drugs is a top priority in the current fight against drug crimes.In this study,196 Raman spectra of seized samples of new psychoactive substances were collected,and the Raman spectral data obtained from five detection instruments were analyzed and identified by combining continuous wavelet transform and convolutional neural network migration learning models.The comparison of the classification efficacy of three deep learning models,VGG16,InceptionV3 and ResNet50,showed that the single-instrument new psychoactive substance recognition ability ranged from 80.2%to 100%,and the combined multi-instrument recognition ability was 88.6%.Through reasonable spectral pre-processing methods,the NPS spectral detection data from different instru-ments can be unified in format and analyzed and identified in batch,which precisely improves the efficiency of data ex-traction and utilization of Raman spectroscopy.

关 键 词：拉曼光谱 光学数据处理 新精神活性物质 迁移学习 连续小波变换 卷积神经网络 

分 类 号：O657.37[理学—分析化学] D918.2[理学—化学]