基于多个太赫兹时域光谱系统的物质识别方法
|
摘要
提出了一种基于多个太赫兹时域光谱系统(THz-TDS)的物质识别方法。将来自不同THz-TDS的光谱数据,通过小波变换去除基线及噪声等干扰信息,并用3次样条插值将不同的采样频率映射到相同的频率上,从而得到标准化后的光谱数据。将此光谱作为支持向量机(SVM)的特征向量,选择合适的核函数并用网格搜索法寻找最优SVM参数,最终得到98.33%的识别准确率。该方法对于准确识别物质具有重要的参考价值。
In this paper, we propose a substance identification method based on multiple terahertz time-domain spectroscopy(THz-TDS) systems. Spectral data from different THz-TDS systems were analyzed by wavelet transform to remove the noise and baseline. Then, the standardized data can be obtained by mapping the sampling frequencies to the same with cubic spline interpolation,which is used as the feature vector of the support vector machine(SVM). Finally, the spectral identification accuracy can be up to 98.33% with an appropriate kernel function and the optimization parameters based on grid search. These results are meaningful for the eventual practical application of exact substance identification.
In this paper, we propose a substance identification method based on multiple terahertz time-domain spectroscopy(THz-TDS) systems. Spectral data from different THz-TDS systems were analyzed by wavelet transform to remove the noise and baseline. Then, the standardized data can be obtained by mapping the sampling frequencies to the same with cubic spline interpolation,which is used as the feature vector of the support vector machine(SVM). Finally, the spectral identification accuracy can be up to 98.33% with an appropriate kernel function and the optimization parameters based on grid search. These results are meaningful for the eventual practical application of exact substance identification.
引文
[1]SIEGEL P H.Terahertz technology in biology and medicine[J].IEEE Transactions on Microwave Theory and Techniques,2004,52(10):2438-2447.
[2]JEPSEN P U,COOKE D G,KOCH M.Terahertz spectroscopy and imaging-Modern techniques and applications[J].Laser&Photonics Reviews,2011,5(1):124-166.
[3]GE H Y,JIANG Y Y,XU Z H,et al.Identification of wheat quality using THz spectrum[J].Optics Express,2014,22(10):12533-12544.
[4]BRIGADA D,ZHANG X C.Chemical identification with information-weighted terahertz spectrometry[J].IEEE Transactions on Terahertz Science and Technology,2012,2(1):107-112.
[5]KING M D,BUCHANAN W D,KORTER T M.Identification and quantification of polymorphism in the pharmaceutical compound diclofenac acid by terahertz spectroscopy and solid-state density functional theory[J].Analytical Chemistry,2011,83(10):3786-3792.
[6]LIU G F,MA X J,MA S H,et al.Identification of explosives 2,4-DNT and 2,6-DNT using terahertz time-domain spectroscopy[J].Chinese Journal of Chemistry,2008,26(7):1257-1261.
[7]CHEN J,CHEN Y Q,ZHAO H W,et al.Absorption coefficients of selected explosives and related compounds in the range of 0.1-2.8 THz[J].Optics Express,2007,15(19):12060-12067.
[8]Cristianini N,Shawe-Taylor J.An Introduction to Support Vector Machines and Other Kernel-based Learning Methods[M].Cambridge:Cambridge University Press,2000.
[9]MARKELZ A G,ROITBERG A,HEILWEIL E J.Pulsed terahertz spectroscopy of DNA,bovine serum albumin and collagen between 0.1 and 2.0 THz[J].Chemical Physics Letters,2000,320(1/2):42-48.
[10]LU J.Parallelizing Mallat algorithm for 2-D wavelet transforms[J].Information Processing Letters,1993,45(5):255-259.
[11]XU L S,ZHANG D,WANG K Q.Wavelet-based cascaded adaptive filter for removing baseline drift in pulse waveforms[J].IEEE Transactions on Biomedical Engineering,2005,52(11):1973-1975.
[12]GREGORY J A,SARFRAZ M.A rational cubic spline with tension[J].Computer Aided Geometric Design,1990,7(1/2/3/4):1-13.
[13]UKIL A.Support vector machine[J].Computer Science,2002,1(4):1-28.
[14]FUREY T S,CRISTIANINI N,DUFFY N,et al.Support vector machine classification and validation of cancer tissue samples using microarray expression data[J].Bioinformatics,2000,16(10):906-914.
[15]FAN R E,CHEN P H,LIN C J.Working set selection using second order information for training Support Vector Machines[J].Journal of Machine Learning Research,2005,6:1889-1918.
[16]CHAPELLE O,VAPNIK V,BOUSQUET O,et al.Choosing multiple parameters for support vector machines[J].Machine Learning,2002,46(1/2/3):131-159.
[17]FENG G.Parameter optimizing for support vector machines classification[J].Computer Engineering and Applications,2011,47(3):123-124,128.
[18]LAMESKI P,ZDRAVEVSKI E,MINGOV R,et al.SVM parameter tuning with grid search and its impact on reduction of model over-fitting[C]//Proceedings of the 15th International Conference on Rough Sets,Fuzzy Sets,Data Mining,and Granular Computing.Cham:Springer,2015.
[19]CORTES C,VAPNIK V.Support-vector networks[J].Machine Learning,1995,20(3):273-297.
[2]JEPSEN P U,COOKE D G,KOCH M.Terahertz spectroscopy and imaging-Modern techniques and applications[J].Laser&Photonics Reviews,2011,5(1):124-166.
[3]GE H Y,JIANG Y Y,XU Z H,et al.Identification of wheat quality using THz spectrum[J].Optics Express,2014,22(10):12533-12544.
[4]BRIGADA D,ZHANG X C.Chemical identification with information-weighted terahertz spectrometry[J].IEEE Transactions on Terahertz Science and Technology,2012,2(1):107-112.
[5]KING M D,BUCHANAN W D,KORTER T M.Identification and quantification of polymorphism in the pharmaceutical compound diclofenac acid by terahertz spectroscopy and solid-state density functional theory[J].Analytical Chemistry,2011,83(10):3786-3792.
[6]LIU G F,MA X J,MA S H,et al.Identification of explosives 2,4-DNT and 2,6-DNT using terahertz time-domain spectroscopy[J].Chinese Journal of Chemistry,2008,26(7):1257-1261.
[7]CHEN J,CHEN Y Q,ZHAO H W,et al.Absorption coefficients of selected explosives and related compounds in the range of 0.1-2.8 THz[J].Optics Express,2007,15(19):12060-12067.
[8]Cristianini N,Shawe-Taylor J.An Introduction to Support Vector Machines and Other Kernel-based Learning Methods[M].Cambridge:Cambridge University Press,2000.
[9]MARKELZ A G,ROITBERG A,HEILWEIL E J.Pulsed terahertz spectroscopy of DNA,bovine serum albumin and collagen between 0.1 and 2.0 THz[J].Chemical Physics Letters,2000,320(1/2):42-48.
[10]LU J.Parallelizing Mallat algorithm for 2-D wavelet transforms[J].Information Processing Letters,1993,45(5):255-259.
[11]XU L S,ZHANG D,WANG K Q.Wavelet-based cascaded adaptive filter for removing baseline drift in pulse waveforms[J].IEEE Transactions on Biomedical Engineering,2005,52(11):1973-1975.
[12]GREGORY J A,SARFRAZ M.A rational cubic spline with tension[J].Computer Aided Geometric Design,1990,7(1/2/3/4):1-13.
[13]UKIL A.Support vector machine[J].Computer Science,2002,1(4):1-28.
[14]FUREY T S,CRISTIANINI N,DUFFY N,et al.Support vector machine classification and validation of cancer tissue samples using microarray expression data[J].Bioinformatics,2000,16(10):906-914.
[15]FAN R E,CHEN P H,LIN C J.Working set selection using second order information for training Support Vector Machines[J].Journal of Machine Learning Research,2005,6:1889-1918.
[16]CHAPELLE O,VAPNIK V,BOUSQUET O,et al.Choosing multiple parameters for support vector machines[J].Machine Learning,2002,46(1/2/3):131-159.
[17]FENG G.Parameter optimizing for support vector machines classification[J].Computer Engineering and Applications,2011,47(3):123-124,128.
[18]LAMESKI P,ZDRAVEVSKI E,MINGOV R,et al.SVM parameter tuning with grid search and its impact on reduction of model over-fitting[C]//Proceedings of the 15th International Conference on Rough Sets,Fuzzy Sets,Data Mining,and Granular Computing.Cham:Springer,2015.
[19]CORTES C,VAPNIK V.Support-vector networks[J].Machine Learning,1995,20(3):273-297.