无机盐FTIR光谱CNN分类识别Inorganic salt FTIR spectrum CNN classification and recognition
金昌达,牟涛涛
摘要(Abstract)：
傅里叶变换红外光谱（Fourier transform infrared spectrum, FTIR）在无机盐鉴定中具有快速、无损分析等优势，但在复杂体系下易受基线漂移与噪声影响，且特征峰重叠导致传统判别方法准确性受限。对此，提出了一种基于一维卷积神经网络（one-dimensional convolutional neural network, 1D-CNN）的无机盐FTIR光谱分类识别方法。首先，选取10种典型无机盐为研究对象，并采用生成对抗网络（generative adversarial network, GAN）进行样本扩增，将数据规模扩展至2 000条，构建FTIR光谱分类识别数据集。其次，构建“波段裁剪—SG(Savitzky-Golay)平滑—非对称最小二乘法（asymmetric least squares, AsLS）基线校正—Min-Max归一化”的预处理流程，以提升光谱质量与一致性。最后，设计一种带有通道注意力机制的1D-CNN，实现端到端分类识别。实验结果表明，该方法的准确率在验证集与测试集上分别达到95.75%与91.00%，表明该方法能够有效提升无机盐FTIR光谱分类性能，为相关快速检测任务提供技术支撑。
关键词(KeyWords)： 卷积神经网络;无机盐;傅里叶变换红外光谱;光谱分类
基金项目(Foundation):
作者(Author): 金昌达,牟涛涛
DOI: 10.16508/j.cnki.11-5866/n.2026.04.012
参考文献(References)：
- [1]周枫然，韩桥，张体强，等.傅里叶变换红外光谱技术的应用及进展[J].化学试剂，2021, 43(8):1001-1009.Zhou Fengran, Han Qiao,Zhang Tiqiang, et al. Application and progress of Fourier transform infrared spectroscopy technology[J]. Chemical Reagents, 2021, 43(8):1001-1009.(in Chinese)
- [2]Ivleva N P. Chemical analysis of microplastics and nanoplastics:challenges,advanced methods,and perspectives[J]. Chemical Reviews,2021,121(19):11886-11936.
- [3]Fine J A,Rajasekar A A,Jethava K P,et al. Spectral deep learning for prediction and prospective validation of functional groups[J]. Chemical Science,2020,11(18):4618-4630.
- [4]Jung G,Jung S G,Cole J M,et al. Automatic materials characterization from infrared spectra using convolutional neural networks[J]. Chemical Science,2023,14(13):3600-3609.
- [5]Dong Mingtan, She Zhenbing, Xiong Xiong, et al. Automated analysis of microplastics based on vibrational spectroscopy:are we measuring the same metrics?[J]. Analytical and Bioanalytical Chemistry, 2022, 414(11):3359-3372.
- [6]Savitzky A,Golay M J E. Smoothing and differentiation of data by simplified least squares procedures[J]. Analytical Chemistry,1964,36(8):1627-1639.
- [7]Ishmukhametov I, Batasheva S, Fakhrullin R. Identification of micro-and nanoplastics released from medical masks using hyperspectral imaging and deep learning[J]. Analyst, 2022,147(20):4616-4628.
- [8]Goodfellow I J,Pouget-Abadie J,Mirza M,et al. Generative adversarial nets[C]//Proceedings of the 28th International Conference on Neural Information Processing Systems. La Jolla,CA, USA:NIPS,2014:2672-2680.
- [9]Lecun Y, Bottou L, Bengio Y,et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE,1998,86(11):2278-2324.
- [10]Michel A P M, Morrison A E,Preston V L,et al. Rapid identification of marine plastic debris via spectroscopic techniques and machine learning classifiers[J]. Environmental Science&Technology,2020,54(17):10630-10637.
- [11]韩卓，刘禹慧，魏晓菲，等.热重分析/红外光谱-气相色谱-质谱联用技术测定松针中6种常见微塑料[J].分析测试学报，2024,43(8):1249-1256.Han Zhuo, Liu Yuhui, Han Xiaofei, et al. Determination of six microplastics in pine needles by TGA/FTIR-GC-MS[J].Journal of Instrumental Analysis, 2024, 43(8):1249-1256.(in Chinese)
- [12]卫辰洁，王继芬，曾啸虎.红外光谱数据融合结合化学计量学无损检测汽车灯罩[J].分析测试学报，2021, 40(7):1043-1048.Wei Chenjie, Wang Jifen, Zeng Xiaohu. Nondestructive detection of automobile lampshades by infrared spectrum data fusion combined with chemometrics[J]. Journal of Instrumental Analysis,, 2021, 40(7):1043-1048.(in Chinese)
- [13]杨青华，林勤保，关伟焰，等.基于傅里叶变换红外光谱和气相色谱-质谱的未知涂料成分分析[J].分析测试学报，2021, 40(11):1571-1579.Yang Qinghua, Lin Qinbao, Guan Weiyan, et al. Analysis of components in unknown coating by Fourier transform infrared spectroscopy and gas chromatography-mass spectrometry[J].Journal of Instrumental Analysis, 2021, 40(11):1571-1579.(in Chinese)
- [14]彭浩然，万希哲，郭乐霞，等.化学计量学在水体微塑料识别中的应用[J].分析测试学报，2025,44(6):1219-1226.Peng Haoran,Wan Xizhe,Guo Lexia,et al. Chemometrics in the identification of microplastics in aquatic ecosystems[J]. Journal of Instrumental Analysis,2025,44(6):1219-1226.(in Chinese)
- [15]谭超，谭成，程斌，等.基于虚拟样本生成的集成模型提升过期药物光谱识别精度[J].分析测试学报，2025,44(6):1131-1138.Tan Chao,Tan Cheng,Cheng Bin,et al. Improving the accuracy of spectral recognition of expired drug by an ensemble model and virtual sample generation[J]. Journal of Instrumental Analysis,2025,44(6):1131-1138.(in Chinese)