|
|
|
|
|
|
Research of Terahertz Time-Domain Spectral Identification Based on Deep Learning |
HU Qi-feng, CAI Jian |
Brainware Terahertz Information Technology Co., Ltd., Hefei 230088, China |
|
|
Abstract Terahertz time-domain spectroscopy (THz-TDS) is an important method for rapid and nondestructive material identification due to its spectral fingerprint properties, which has broad application exploitation in the nondestructive inspection of drugs and explosives. Spectral identification is one of the most important aspects of the applied research of THz-TDS. Most existing spectral identification methods are machine-learning based classification of manually selected features or thresholding classification of absorption spectral peak. Those methods are not adapt well to low signal-to-noise ratio, because some materials have few or no spectral absorption peaks features in the terahertz waveband and spectra are affected easily by concentrations of samples, air humidity and noises. Meanwhile computational cost increases with data quantity and category. In recent years, with the rise of deep learning technology, the methods represented by CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network) have been widely applied to fields such as computer vision and natural language processing where they have been shown to produce better results than traditional machine learning methods. Due to the strong nonlinear classification capability of deep learning technology, two networks respectively were designed based on RNN and CNN for spectral identification in this paper: one-dimensional spectral line classifier based on RNN and two-dimensional spectral image classifier based on CNN. To simulate the practical application scenario, over 20 000 terahertz time-domain spectra of 12 materials were measured in a non-vacuum environment as training-set and test-set. After analyzing the effects of concentrations of samples and air humidity on spectra, S-G(Savitzky-Golay) filter was introduced to reduce noises of spectra. Experimental results show that S-G filter could improve the identification accuracy, because processed spectra have more obvious feature compared with the unprocessed spectra; the proposed methods based on RNN and CNN are more accurate and faster on the test-set, compared with traditional machine learning algorithm k-NN (k-Nearest Neighbor); CNN demonstrated better robustness to noises than RNN on spectral identification task. Therefore, deep learning technology could be utilized for quick and effective identification terahertz time-domain spectra, which provide a theoretical and experimental basis for new nondestructive safety inspection techniques.
|
Received: 2019-11-15
Accepted: 2020-03-12
|
|
|
[1] Trofimov V A, Varentsova S A. PLOS ONE, 2018, 13(8): e0201297.
[2] Liu J, Li Z, Hu F, et al. Optical and Quantum Electronics, 2015, 47(2): 313.
[3] Yin X, Mo W, Wang Q, et al. Advances in Condensed Matter Physics, 2018, 2018: 1618750.
[4] Liang J, Guo Q, Chang T, et al. Optik, 2018, 174: 7.
[5] Yin M, Tang S, Tong M. Analytical Methods, 2016, 8(13): 2794.
[6] Mumtaz M, Mahmood A, Khan S D, et al. Applied Spectroscopy, 2017, 71(3): 456.
[7] XIE Qi, YANG Hong-ru, LI Hong-guang, et al(解 琪, 杨鸿儒, 李宏光, 等). Optics and Precision Engineering(光学精密工程), 2016,24(10): 2392.
[8] Naftaly M. IEEE Sensors Journal, 2013, 13(1): 8.
[9] LI Jin, LIU Quan-cheng, XIONG Liang(李 进, 刘泉澄, 熊 亮). Laser & Optoelectronics Progress(激光与光电子学进展), 2018, 55(9): 70.
[10] Liu H, Zhang Z, Yang Y, et al. Optik, 2018, 172: 668.
[11] Yu S, Jia S, Xu C. Neurocomputing, 2017, 219: 88.
[12] Maggiori E, Tarabalka Y, Charpiat G, et al. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(2): 645.
[13] Karim F, Majumdar S, Darabi H, et al. IEEE Access, 2018, 6: 1662. |
[1] |
LI Jie, ZHOU Qu*, JIA Lu-fen, CUI Xiao-sen. Comparative Study on Detection Methods of Furfural in Transformer Oil Based on IR and Raman Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 125-133. |
[2] |
LI Xin-ting, ZHANG Feng, FENG Jie*. Convolutional Neural Network Combined With Improved Spectral
Processing Method for Potato Disease Detection[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 215-224. |
[3] |
LAN Yan1,WANG Wu1,XU Wen2,CHAI Qin-qin1*,LI Yu-rong1,ZHANG Xun2. Discrimination of Planting and Tissue-Cultured Anoectochilus Roxburghii Based on SMOTE and Inception-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 158-163. |
[4] |
WAN Mei, ZHANG Jia-le, FANG Ji-yuan, LIU Jian-jun, HONG Zhi, DU Yong*. Terahertz Spectroscopy and DFT Calculations of Isonicotinamide-Glutaric Acid-Pyrazinamide Ternary Cocrystal[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(12): 3781-3787. |
[5] |
WANG Qi-biao1, HE Yu-kai1, LUO Yu-shi1, WANG Shu-jun1, XIE Bo2, DENG Chao2*, LIU Yong3, TUO Xian-guo3. Study on Analysis Method of Distiller's Grains Acidity Based on
Convolutional Neural Network and Near Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(12): 3726-3731. |
[6] |
HUANG Meng-qiang1, KUANG Wen-jian2, 3*, LIU Xiang1, HE Liang4. Quantitative Analysis of Cotton/Polyester/Wool Blended Fiber Content by Near-Infrared Spectroscopy Based on 1D-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(11): 3565-3570. |
[7] |
CHEN Jia-wei1, 2, ZHOU De-qiang1, 2*, CUI Chen-hao3, REN Zhi-jun1, ZUO Wen-juan1. Prediction Model of Farinograph Characteristics of Wheat Flour Based on Near Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(10): 3089-3097. |
[8] |
LI Yang1, LI Xiao-qi1, YANG Jia-ying1, SUN Li-juan2, CHEN Yuan-yuan1, YU Le1, WU Jing-zhu1*. Visualisation of Starch Distribution in Corn Seeds Based on Terahertz Time-Domain Spectral Reflection Imaging Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(09): 2722-2728. |
[9] |
CAI Jian-rong1, 2, HUANG Chu-jun1, MA Li-xin1, ZHAI Li-xiang1, GUO Zhi-ming1, 3*. Hand-Held Visible/Near Infrared Nondestructive Detection System for Soluble Solid Content in Mandarin by 1D-CNN Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(09): 2792-2798. |
[10] |
PU Shan-shan, ZHENG En-rang*, CHEN Bei. Research on A Classification Algorithm of Near-Infrared Spectroscopy Based on 1D-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(08): 2446-2451. |
[11] |
TANG Ting, PAN Xin*, LUO Xiao-ling, GAO Xiao-jing. Fusion of ConvLSTM and Multi-Attention Mechanism Network for
Hyperspectral Image Classification[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(08): 2608-2616. |
[12] |
LI Wen-xia1, DU Yu-jun2, WANG Yue1, LIU Zheng-dong3*, ZHENG Jia-hui1, DU Wen-qian1, WANG Hua-ping4. Research on On-Line Efficient Near-Infrared Spectral Recognition and Automatic Sorting Technology of Waste Textiles Based on Convolutional Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(07): 2139-2145. |
[13] |
LIANG Wan-jie1, FENG Hui2, JIANG Dong3, ZHANG Wen-yu1, 4, CAO Jing1, CAO Hong-xin1*. Early Recognition of Sclerotinia Stem Rot on Oilseed Rape by Hyperspectral Imaging Combined With Deep Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(07): 2220-2225. |
[14] |
JIN Cheng-liang1, WANG Yong-jun2*, HUANG He2, LIU Jun-min3. Application of High-Dimensional Infrared Spectral Data Preprocessing in the Origin Identification of Traditional Chinese Medicinal Materials[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(07): 2238-2245. |
[15] |
LI Hao-dong1, 2, LI Ju-zi1*, CHEN Yan-lin1, HUANG Yu-jing1, Andy Hsitien Shen1*. Establishing Support Vector Machine SVM Recognition Model to Identify Jadeite Origin[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(07): 2252-2257. |
|
|
|
|