|
|
|
|
|
|
| Research on the Recognition of Mixed Mid-Infrared Spectra of Hazardous Chemicals Based on an Improved High-Order Residual Network |
| FAN Bing-rui1, 2, ZHAI Ai-ping1, WANG Dong1, LIANG Ting3, ZHANG Gen-wei2*, CAO Shu-ya2* |
1. College of Physics and Optoelectronics, Taiyuan University of Technology, Taiyuan 030024, China
2. National Key Laboratory of Nuclear, Biological and Chemical Hazard Protection, Beijing 102205, China
3. Institute of Chemical Defense, Beijing 102205, China
|
|
|
|
|
Abstract Amid the rapid development of the chemical industry, the large-scale production and use of hazardous chemicals have brought significant environmental risks and potential threats to human health. Consequently, there is an urgent need to develop efficient monitoring and identification methods to address these challenges. Mid-infrared (MIR) spectroscopy, characterized by its unique molecular “fingerprint” recognition capability and high sensitivity, has been widely applied in the identification and analysis of hazardous chemicals. Meanwhile, recent breakthroughs in deep learning, particularly in feature extraction and object detection, have provided novel solutions for addressing complex data analysis problems. To meet the need for precise identification of hazardous chemicals under complex background conditions, this study proposes a novel model that combines Depthwise Separable Convolution (DSC) and a High-Order Residual Network (HORN), referred to as DSC-HORN. This model is designed to enhance further the efficiency of feature extraction and the accuracy of recognition for complex MIR spectral data. By leveraging depthwise separable convolution, the DSC-HORN model decomposes standard convolutions into depthwise and pointwise convolutions, resulting in a reduction of approximately 65.89% in parameter count and computational complexity. This optimization increases the operational speed of the DSC-HORN model by approximately four times compared to traditional one-dimensional convolutional neural networks (1D-CNNs), while significantly reducing memory consumption and enhancing its adaptability in resource-constrained environments. For the experiments, this study collected MIR spectral data of a 10% DMMP solution mixed with 11 types of background materials. After data preprocessing and removal of outlier samples, a total of 528 valid spectral datasets were obtained for model construction. To ensure balanced distribution of different labeled samples across subsets and to accurately reflect the overall data characteristics, the dataset was divided into training, validation, and testing sets using a stratified random sampling method, with a 6∶1∶3 ratio for training, validation, and testing, respectively. The experimental results demonstrated that the DSC-HORN model exhibits significant advantages in recognition efficiency and accuracy compared to existing models. Compared to traditional models such as K-Nearest Neighbors(KNN), Random Forest(RF), Robust Mode Regression Sparse Non-negative Matrix Factorization(MR-NMF), and Backpropagation Neural Networks(BP), as well as deep learning models like1D-Convolutional Neural Network(1D-CNN), the DSC-HORN model achieved a recognition accuracy of 96.84%. This represents improvements of 5.7%, 13.93%, 5.07%, 10.76%, and 6.33%, respectively, over the models above. These performance enhancements can be attributed to the parameter optimization achieved through depthwise separable convolution and the efficient feature extraction enabled by high-order residual connections. The results further confirm that the DSC-HORN model is not only an efficient and accurate MIR hazardous chemical recognition model but also provides a novel technical pathway for real-time monitoring and precise identification of hazardous chemicals. Additionally, the MIR recognition technology, based on the improved high-order residual network, has great potential to promote further the widespread application of MIR spectroscopy in complex scenarios.
|
|
Received: 2024-12-05
Accepted: 2025-05-02
|
|
|
|
Corresponding Authors:
ZHANG Gen-wei, CAO Shu-ya
E-mail: zhgw1984@163.com; caoshuya@skinbcpc.cn
|
|
[1] Zhu L, Lu S H, Zhang Y H, et al. Vibrational Spectroscopy, 2020, 109: 103071.
[2] Zhang L J, Huang X D, Fan X M, et al. Environmental Technology, 2021, 42(2): 270.
[3] Ceniti C, Spina A A, Piras C, et al. Foods, 2023, 12: 2917.
[4] De Bruyne S, Speeckaert M M, Delanghe J R. Critical Reviews in Clinical Laboratory Sciences, 2018, 55(1): 1.
[5] Wang J, Niu X, Zheng L, et al. Sensors, 2016, 16: 1941.
[6] GUO Teng-xiao, DING Xue-quan, DONG Xiao-qiang, et al（郭腾霄，丁学全，董晓强，等）. Computers and Applied Chemistry(计算机与应用化学), 2013, 30(1): 18.
[7] LENG Si-yu, QIAO Jia-hui, WANG Lian-qing, et al（冷思雨，乔嘉慧，王连庆，等）. Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2023, 43(1): 303.
[8] Lee L C, Liong C Y, Jemain A A. Chemometrics and Intelligent Laboratory Systems, 2018, 182: 90.
[9] Jin X, Che J, Chen Y. IEEE Access, 2021, 9: 10940.
[10] Nassif A B, Shahin I, Attili I, et al. IEEE Access, 2019, 7: 19143.
[11] Min S, Lee B, Yoon S. Briefings in Bioinformatics, 2017, 18(5): 851.
[12] Kuutti S, Bowden R, Jin Y, et al. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(2): 712.
[13] Srivastava H, Sarawadekar K. Proceedings of 2020 IEEE Applied Signal Processing Conference (ASPCON), Kolkata, India, 2020, 1.
[14] Ruuska S, Hämäläinen W, Kajava S, et al. Behavioural Processes, 2018, 148: 56.
|
| [1] |
ZHU Yi-feng1, 3, LIU Jin-peng1, 3, SONG Zheng-xun1, 3*, ZHOU Xiao-jun4*, ZHOU Mei5, REN Jiao-jiao2, 3, YANG Wen-tao2, CUI Zong-yu2, 3. Study of Differential Diagnosis of Early NPC and Nasopharyngitis Based on THz Spectra[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 941-946. |
| [2] |
LI Xin-yi1, KONG De-ming1*, NING Xiao-dong2, CUI Yao-yao3. Research on Emulsified Oil Spill Detection Methods Based on
Mid-Infrared Spectroscopy Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(03): 631-636. |
| [3] |
WANG Lei1, 2, CHEN Yuan-jie1, 3, LI Lei4, LIU Yong-hong1, 2, XU Ke-ke1, 2, YU Huan-ying1, YANG Lin-lin1, 2, DONG Cheng-ming1, 2, QIAO Lu1, 2*. Rapid Identification of Corni Frucutus From Different Habitats Based on Mid-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(03): 761-767. |
| [4] |
MA Hu-yishan1, 2, PAN Nan2, LIN Zhen-yu3, CHEN Xiao-ting2, WU Jing-na4, ZHANG Fang1*, LIU Zhi-yu2*. Research Progress on the Vibrational Spectroscopy Technology in the Quality Detection of Fish Oil[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(02): 301-311. |
| [5] |
XIAO Zhong-liang, YUAN Rong-yao, FU Zhuang, LIU Cheng, YIN Bi-lu, XIAO Min-zhi, ZHAO Ting-ting, KUANG Yin-jie, SONG Liu-bin*. Study on the Aging Behavior of Transformer Oil Based on Machine
Learning and Infrared Spectroscopy Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(02): 434-442. |
| [6] |
QIAO Lu1, 2, LIU Yong-hong1, 2, XU Ke-ke1, 2, YU Huan-ying1, CHEN Yuan-jie3, YANG Lin-lin1, 2, DONG Cheng-ming1, 2*, WANG Lei1*. Analysis of Mid-Infrared Spectral Characteristics of Soils Cultivated With Salvia Miltiorrhiza at Different Intervals Based on Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(02): 483-491. |
| [7] |
YANG Cheng-en1, 2, GUO Rui-xue1, 3, XIN Ming-hao2, LI Meng4, LI Yu-ting2*, SU Ling1, 3*. Quantitative Determination of Polyphenols in Aronia Melanocarpa (Michx.) Elliott. by Mid-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3075-3081. |
| [8] |
SUN Yu-tong1, 2, LÜ Shu-qiang1, 2, HOU Miao-le1, 2*. A Mixed Pigment Identification Method Based on Spectra Interval[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(08): 2357-2364. |
| [9] |
XU Jian-kang1, LIU Bing-xin1, 2*, DU Yu-long1, LI Ying1, 2, LIU Peng1, 2, CHEN Peng1, 2. Visible Near-Infrared Reflection Spectrum Characteristics and Angular
Effects of Sea Ice Contaminated by Ship Oil Spills[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(07): 2075-2082. |
| [10] |
XIA Yan-qiu1, XIE Pei-yuan1, NAY MIN AUNG1, ZHANG Tao1, FENG Xin1, 2*. The Improved Genetic Algorithm is Embedded Into the Classical
Classification Algorithm to Realize the Synchronous
Identification of Small Quantity and Multi Types of
Lubricating Oil Additives[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(03): 744-750. |
| [11] |
KANG Rui1, 2, CHENG Ya-wen1, 2, ZHOU Ling-li1, 2, REN Ni1, 2*. A Novel Classification Method of Foodborne Bacterial Species Based on Hyperspectral Microscopy Imaging Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(02): 392-397. |
| [12] |
YANG Cheng-en1, 2, LI Meng3, LU Qiu-yu2, WANG Jin-ling4, LI Yu-ting2*, SU Ling1*. Fast Prediction of Flavone and Polysaccharide Contents in
Aronia Melanocarpa by FTIR and ELM[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 62-68. |
| [13] |
DUAN Ming-xuan1, LI Shi-chun1, 2*, LIU Jia-hui1, WANG Yi1, XIN Wen-hui1, 2, HUA Deng-xin1, 2*, GAO Fei1, 2. Detection of Benzene Concentration by Mid-Infrared Differential
Absorption Lidar[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(11): 3351-3359. |
| [14] |
LIU Bo-yang1, GAO An-ping1*, YANG Jian1, GAO Yong-liang1, BAI Peng1, Teri-gele1, MA Li-jun1, ZHAO San-jun1, LI Xue-jing1, ZHANG Hui-ping1, KANG Jun-wei1, LI Hui1, WANG Hui1, YANG Si2, LI Chen-xi2, LIU Rong2. Research on Non-Targeted Abnormal Milk Identification Method Based on Mid-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(10): 3009-3014. |
| [15] |
WANG Yu-ye1, 2, LI Hai-bin1, 2, JIANG Bo-zhou1, 2, GE Mei-lan1, 2, CHEN Tu-nan3, FENG Hua3, WU Bin4ZHU Jun-feng4, XU De-gang1, 2, YAO Jian-quan1, 2. Terahertz Spectroscopic Early Diagnosis of Cerebral Ischemia in Rats[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(03): 788-794. |
|
|
|
|
