基于1D-CNN的生物气溶胶衰减全反射傅里叶变换红外光谱识别  

Attenuated Total Reflection Fourier Transform Infrared Spectral Identification of Bioaerosol Based on 1D-CNN

作　　者：汪洋[1,2] 童晶晶 李相贤 韩昕 秦玉胜 方仁杰 高闽光 Wang Yang;Tong Jingjing;Li Xiangxian;Han Xin;Qin Yusheng;Fang Renjie;Gao Minguang(University of Science and Technology of China,Hefei 230026,Anhui,China;Key Laboratory of Environmental Optics and Technology,Anhui Institute of Optics and Fine Mechanics,Hefei Institutes of Physical Science,Chinese Academy of Sciences,Hefei 230031,Anhui,China)

机构地区：[1]中国科学技术大学,安徽合肥230026 [2]中国科学院合肥物质科学研究院安徽光学精密机械研究所,中国科学院环境光学与技术重点实验室,安徽合肥230031

出　　处：《光学学报》2024年第24期301-307,共7页Acta Optica Sinica

基　　金：国家自然科学基金(42075135);国家重点研发计划(2022YFB2602000,2022YFC3700500)。

摘　　要：针对环境监测和公共健康保护中生物气溶胶快速识别的需求,结合一维卷积神经网络(1DCNN)和衰减全反射傅里叶变换红外(ATRFTIR)光谱技术,提出了一种生物气溶胶识别方法。通过小波包变换和Savitzky Golay(SG)卷积平滑算法对ATRFTIR光谱数据进行预处理,利用1DCNN模型进行深度特征提取和分类。与传统的支持向量机(SVM)方法相比,所建模型在识别6种常见生物气溶胶样本上展现出了99.9%以上的识别准确率,显著优于SVM方法的95%。此外,通过交叉验证和低含量样品数据的验证实验,1DCNN模型展现出了高稳定性和良好的泛化能力。证实了将1DCNN与ATRFTIR光谱技术相结合,可以实现生物气溶胶的快速、准确识别,为生物污染事件的快速响应提供了有效的技术支持。Objective As an important component of the atmospheric environment,bioaerosols have a profound effect on environmental quality,climate change,and human health.As environmental and public health problems intensify,the monitoring and identification of bioaerosols have attracted widespread attention.However,traditional bioaerosol identification methods,such as microbial culture and molecular biology techniques,are slow and complex.We combine attenuated total reflection Fourier transform infrared(ATRFTIR)spectroscopy with onedimensional convolutional neural network(1DCNN)to leverage the high sensitivity,noninvasive and realtime advantages of spectroscopic technology,as well as deep learning powerful capabilities in feature extraction and classification of complex spectral data,and build an efficient and accurate bioaerosol identification model.Methods Bioaerosol samples,including three types of bacteria and three types of fungi,are used as the research object,and highquality infrared absorption spectrum data are collected using a Fourier transform infrared spectrometer with an attenuated total reflection(ATR)accessory.To improve data quality,preprocessing techniques such as wavelet packet transform and SavitzkyGolay filtering are used for baseline correction and noise filtering.On this basis,a 1DCNN model,including a convolution layer,a pooling layer,a dropout layer,and a fully connected layer,is constructed to utilize its powerful feature extraction and classification capabilities for the fast and accurate identification of bioaerosols.The effectiveness and superiority of the model are fully verified through reasonable data set division,multiangle performance evaluation,and comparison with traditional machine learning methods.A mixed sample test plan of different concentrations is designed to further evaluate the model's generalization ability in complex environments.Results and Discussions Through comparative analysis of test set recognition accuracy,the 1DCNN model proposed in this paper performs exceptionally wel

关 键 词：光谱学 生物气溶胶 光谱识别 傅里叶变换红外光谱技术 卷积神经网络 

分 类 号：O657.33[理学—分析化学]