|
|
|
|
|
|
Terahertz Spectroscopic Identification with Diffusion Maps |
倪家鹏1,沈 韬1, 2* ,朱 艳2,李灵杰1,毛存礼1,余正涛1 |
1. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, China
2. Faculty of Materials Science and Engineering, Kunming University of Science and Technology, Kunming 650093, China |
|
|
Abstract Feature extraction is the key issue for the identification of terahertz spectroscopy. For the traditional method, it is identified by different absorption peaks as the features that extracted through manual method. However, for many materials,there are no apparent spectral graphics features in the terahertz band, such as peaks, valleys and etc. To this end, the researchers reduce the dimension from the high-dimensional terahertz spectroscopy data and extract the features through statistics learning and machine learning methods. Linear method is easy to cause greater error due to the nonlinear nature of terahertz spectroscopy data, especially when different materials of spectrum curves are very similar. To address this issue, a novel terahertz spectroscopy identification approach with Diffusion Maps (DM) was studied in this paper. Diffusion Maps can realize nonlinear dimensionality reduction while maintaining the internal geometry of the data. In addition, the manifold features extracted by the method have good discrimination and clustering performance. Firstly, S-G filter and cubic spline interpolation were used to smooth and uniform the resolution of terahertz transmission spectra of ten kinds of substances in the same frequency band. Secondly, high-dimensional data of terahertz spectra is mapped to the low-dimensional feature space by using DM so that we can extract the manifold features of terahertz spectroscopy. Finally, a Multi-class Support Vector Machine (M-SVM) classifier is applied to classify these terahertz spectra. Experimental results show that, compared with Principal Component Analysis (PCA) and Isometric Mapping (ISOMAP), manifold features of terahertz spectroscopy extracted by DM have higher degree of differentiation. Besides, DM can get the estimation of intrinsic dimension of terahertz spectra directly. So this proposed method provides a novel approach to identify similar terahertz spectrum quickly and accurately.
|
Received: 2016-07-15
Accepted: 2016-11-08
|
|
Corresponding Authors:
SHEN Tao
E-mail: shentao@kmust.edu.cn
|
|
[1] WANG Guo, WANG Wei-ning(王 果, 王卫宁). Acta Physico-Chimica Sinica(物理化学学报), 2012, 28(7): 1579.
[2] MENG Zeng-rui, ZHANG Wei-bin, DU Yu(孟增睿, 张伟斌, 杜 宇). Acta Physica Sinica(物理学报), 2015, 64(7): 214.
[3] Qin J, Xie L, Ying Y. Analytical Chemistry, 2014, 86(23): 11750.
[4] ZHU Si-yuan, ZHANG Man, SHEN Jing-ling(朱思原, 张 曼, 沈京玲). Infrared and Laser Engineering(红外与激光工程), 2013, 42(3): 626.
[5] LI Li-long, XIANG Yang, WU Lei(李利龙, 向 洋, 吴 磊). High Power Laser and Particle Beams(强激光与粒子束), 2013, 25(6): 1566.
[6] Zhang W, Nie J, Tu S. Optical and Quantum Electronics, 2015, 47: 3533.
[7] Zhan H, Wu S, Bao R. Fuel, 2015, 143: 189.
[8] Coifman R R, Lafon S. Applied and Computational Harmonic Analysis, 2006, 21(1): 5.
[9] Luca Rossi, Andrea Torsello, Edwin R Hancock. Pattern Recognition, 2015, 48(11): 3357.
[10] HUANG Yun-juan, LI Fan-zhang(黄运娟, 李凡长). Journal of Software(软件学报), 2013, 24(11): 2656.
[11] Izquierdo-Verdiguier E, Jenssen R, Gomez-Chova L. Neurocomputing, 2015, 149(5): 1299.
[12] Wang L, Wang K, Li R. Iet Computer Vision, 2015, 9(5): 655.
[13] Horvath D, Ulicny J, Brutovsky B. Connection Science, 2016,(1): 1.
[14] Lisboa P J G, Martin-Guerrero J D, Vellido A. Expert Systems with Applications, 2015, 42(22): 8982.
|
[1] |
NIE Mei-tong1,2, XU De-gang1,2*, WANG Yu-ye1,2*, TANG Long-huang1,2, HE Yi-xin1,2, LIU Hong-xiang1,2, YAO Jian-quan1,2. Investigation on Characteristics of Edible Oil Spectra with Terahertz Time-Domain Attenuated Total Reflection Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(07): 2016-2020. |
[2] |
CHEN Xi-ai1,2,3, WU Xue1, ZHANG Song1, WANG Ling1. Study of Plant Growth Regulators Detection Technology Based on Terahertz Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(03): 665-669. |
[3] |
YU Jiang-ping1, LI Chun2, YI Wen-ting1, JIN Biao-bing2, LIU Yun-fei1, JIANG Ling1*. Temperature Dependence of Terahertz Spectra of Amino Acid[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(02): 386-393. |
[4] |
MA Pin1, YANG Yu-ping1*, ZHANG Zhen-wei2, ZHANG Cun-lin2. Study on the FTIR and THz Spectra of Azitromycin Drugs with Different Physical Forms[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(10): 3016-3020. |
[5] |
HUANG Yu1, SUN Ping1*, ZHANG Zheng1, JIN Chen1, LIU Wei2, WANG Wen-ai2. Experimental Study on Terahertz Spectrum of Lacidipine
[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(08): 2356-2359. |
[6] |
LIANG Liang1, TONG Min-ming1*, WEI Ming-sheng1, 2, LI Meng1, DONG Hai-bo1. Research on Scattering in the THz Quantitative Analysis of Granulated Coal[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(08): 2365-2370. |
[7] |
DAI Ze-lin, XU Xiang-dong*, GU Yu, ZOU Rui-jiao, HAN Shou-sheng, PENG Yong, LIAN Yu-xiang, WANG Fu, LI Xin-rong, CHEN Zhe-geng, SUN Ming-hui, JIANG Ya-dong. Experimental and Theoretical Study on Terahertz Spectra for Regenerated Cellulose [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(03): 697-703. |
[8] |
ZHANG Liang-liang1, ZHANG Rui2, HUANG Su-xia1, ZHANG Cun-lin1 . Terahertz Spectroscopic Investigation of Substrate Materials for Biological Application in the Frequency Range of 1~15 THz [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(02): 346-349. |
[9] |
CHEN Tao, LI Zhi, HU Fang-rong, YIN Xian-hua, XU Chuan-pei* . Discrimination of GMOs Using Terahertz Spectroscopy and CS-SVM [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(02): 618-623. |
[10] |
WANG Fang1, 4, ZHAO Dong-bo2, JIANG Ling1, XU Li3, SUN Hai-jun3, LIU Yun-fei1* . Research on THz and Raman Spectra of RNA Nucleobases [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(12): 3863-3869. |
[11] |
LI Jia-yu1, SUN Ping1*, ZOU Yun1, LIU Wei2, WANG Wen-ai2 . Comparison and Analysis of Iterative and Genetic Algorithms Used to Extract the Optical Parameters of Glucose Polycrystalline[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(12): 3875-3880. |
[12] |
MA Yu-ting1,2, ZHANG Jin1, SHI Chang-cheng2*, LI Wei2, HAN Xiao-hui1, PENG Xiao-yu2, WEI Dong-shan2, DU Chun-lei2, CUI Hong-liang1,2*. Current Status and Recent Advances in Research and Application of THz Technology in Articular Cartilage Detection [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(07): 2031-2035. |
[13] |
DONG Jia-meng1, 2, PENG Xiao-yu2*, MA Xiao-hui1*, LIU Yi1, 2, WEI Dong-shan2, CUI Hong-liang2, DU Chun-lei2. Progress of Detection Technology of Ultra-Broadband THz Time-Domain Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(05): 1277-1283. |
[14] |
ZHANG Liang-liang1, ZHANG Rui2, XU Xiao-yan1, ZHANG Cun-lin1 . The Detection of Ultra-Broadband Terahertz Spectroscopy of InP Wafer by Using Coherent Heterodyne Time-Domain Spectrometer [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(02): 322-325. |
[15] |
MA Shuai1, SHEN Tao1, 2*, WANG Rui-qi1, LAI Hua1, YU Zheng-tao1 . Terahertz Spectroscopic Identification with Deep Belief Network [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(12): 3325-3329. |
|
|
|
|