基于差分-主成分分析-支持向量机的有机化合物太赫兹吸收光谱识别方法
|
摘要
针对有机化合物的太赫兹时域光谱数据,提出了一种基于差分-主成分分析(PCA)-支持向量机(SVM)的有机化合物识别方法。基于物质样本的太赫兹时域信号计算得到太赫兹吸收光谱,对0.2～2.5 THz频率区间内的数据进行特征提取。在特征提取中,提出了基于差分数据的样本容量扩充方法,并结合PCA进行了特征的提取。利用SVM建立了提取的特征与物质类别对应关系的数学模型,并根据建立的模型对未知样本进行了识别研究。利用所提方法对15种有机化合物的太赫兹光谱数据进行了识别,正确识别率为93.33%。将所提方法与线性判别分析法及吸收峰频率-幅值法进行了对比,结果表明基于差分-PCA-SVM的有机化合物识别方法的正确识别率最高。
This paper proposes a method for identifying organic compounds by applying a differential principal-component-analysis(PCA)-support-vector-machine(SVM) to the terahertz time-domain spectral data. First, the terahertz absorption spectrum is calculated according to the terahertz time-domain signal of the material sample; then, the features of the data in the frequency range of 0.2-2.5 THz are extracted. During the feature extraction, an expansion-of-sample-size method based on differential data is proposed and combined with the PCA method to achieve the feature extraction. Finally, the SVM is used to establish a mathematical model for the corresponding relationship between the extracted features and the material category, and the unknown samples are identified according to this model. The terahertz-spectral data of 15 organic compounds are identified using the proposed method, and the correct recognition rate is 93.33%. The experimental results show that the correct recognition rate of organic compounds by the proposed method is the highest when compared with those by the linear-discriminant analysis method and the absorption peak frequency-amplitude method.
This paper proposes a method for identifying organic compounds by applying a differential principal-component-analysis(PCA)-support-vector-machine(SVM) to the terahertz time-domain spectral data. First, the terahertz absorption spectrum is calculated according to the terahertz time-domain signal of the material sample; then, the features of the data in the frequency range of 0.2-2.5 THz are extracted. During the feature extraction, an expansion-of-sample-size method based on differential data is proposed and combined with the PCA method to achieve the feature extraction. Finally, the SVM is used to establish a mathematical model for the corresponding relationship between the extracted features and the material category, and the unknown samples are identified according to this model. The terahertz-spectral data of 15 organic compounds are identified using the proposed method, and the correct recognition rate is 93.33%. The experimental results show that the correct recognition rate of organic compounds by the proposed method is the highest when compared with those by the linear-discriminant analysis method and the absorption peak frequency-amplitude method.
引文
[1] Cheng D Z.Controllability of switched bilinear systems[J].IEEE Transactions on Automatic Control,2005,50(4):511-515.
[2] Rudd J V,Johnson J L,Mittleman D M.Quadrupole radiation from terahertz dipole antennas[J].Optics Letters,2000,25(20):1556-1558.
[3] Chen T,Zhang C J,Xu C P.Study onterahertz time-domain spectroscopy of tartaric acid isomers[J].Laser & Optoelectronics Progress,2017,54(8):081202.陈涛,张超杰,许川佩.酒石酸异构体的太赫兹时域光谱研究[J].激光与光电子学进展,2017,54(8):081202.
[4] Trofimov V,Varentsova S.An effective method for substance detection using the broad spectrum THz signal:a “terahertz nose”[J].Sensors,2015,15(6):12103-12132.
[5] Ni J P,Shen T,Zhu Y,et al.Terahertz spectroscopic identification with diffusion maps[J].Spectroscopy and Spectral Analysis,2017,37(8):2360-2364.倪家鹏,沈韬,朱艳,等.基于扩散映射的太赫兹光谱识别[J].光谱学与光谱分析,2017,37(8):2360-2364.
[6] Liu K,Li B,Zeng X X,et al.Method of terahertz time-domain spectroscopy classification based on mani-fold learning and support vector machine[J].Computer Engineering and Applications,2015,51(24):141-144,175.刘坤,李飚,曾祥鑫,等.基于流形学习和支持向量机的太赫兹谱分类[J].计算机工程与应用,2015,51(24):141-144,175.
[7] Xie Q,Yang H R,Li H G,et al.Explosive identification based on terahertz time-domain spectral system[J].Optics and Precision Engineering,2016,24(10):2392-2399.解琪,杨鸿儒,李宏光,等.基于太赫兹时域光谱系统的爆炸物识别[J].光学精密工程,2016,24(10):2392-2399.
[8] Wang K,Wang H Q,Yin Y,et al.Pigment spectral matching recognition method based on adaptive edit distance[J].Laser & Optoelectronics Progress,2018,55(11):113004.王可,王慧琴,殷颖,等.基于自适应编辑距离的颜料光谱匹配识别方法[J].激光与光电子学进展,2018,55(11):113004.
[9] Zhang W T,Li Y W,Zhan P P,et al.Recognition of transgenic soybean oil based on terahertz time- domain spectroscopy and PCA-SVM[J].Infrared and Laser Engineering,2017,46(11):1125004.张文涛,李跃文,占平平,等.基于太赫兹时域光谱技术与PCA-SVM的转基因大豆油鉴别研究[J].红外与激光工程,2017,46(11):1125004.
[10] Pohl A,De?mann N,Dutzi K,et al.Identification of unknown substances by terahertz spectroscopy and multivariate data analysis[J].Journal of Infrared,Millimeter,and Terahertz Waves,2016,37(2):175-188.
[2] Rudd J V,Johnson J L,Mittleman D M.Quadrupole radiation from terahertz dipole antennas[J].Optics Letters,2000,25(20):1556-1558.
[3] Chen T,Zhang C J,Xu C P.Study onterahertz time-domain spectroscopy of tartaric acid isomers[J].Laser & Optoelectronics Progress,2017,54(8):081202.陈涛,张超杰,许川佩.酒石酸异构体的太赫兹时域光谱研究[J].激光与光电子学进展,2017,54(8):081202.
[4] Trofimov V,Varentsova S.An effective method for substance detection using the broad spectrum THz signal:a “terahertz nose”[J].Sensors,2015,15(6):12103-12132.
[5] Ni J P,Shen T,Zhu Y,et al.Terahertz spectroscopic identification with diffusion maps[J].Spectroscopy and Spectral Analysis,2017,37(8):2360-2364.倪家鹏,沈韬,朱艳,等.基于扩散映射的太赫兹光谱识别[J].光谱学与光谱分析,2017,37(8):2360-2364.
[6] Liu K,Li B,Zeng X X,et al.Method of terahertz time-domain spectroscopy classification based on mani-fold learning and support vector machine[J].Computer Engineering and Applications,2015,51(24):141-144,175.刘坤,李飚,曾祥鑫,等.基于流形学习和支持向量机的太赫兹谱分类[J].计算机工程与应用,2015,51(24):141-144,175.
[7] Xie Q,Yang H R,Li H G,et al.Explosive identification based on terahertz time-domain spectral system[J].Optics and Precision Engineering,2016,24(10):2392-2399.解琪,杨鸿儒,李宏光,等.基于太赫兹时域光谱系统的爆炸物识别[J].光学精密工程,2016,24(10):2392-2399.
[8] Wang K,Wang H Q,Yin Y,et al.Pigment spectral matching recognition method based on adaptive edit distance[J].Laser & Optoelectronics Progress,2018,55(11):113004.王可,王慧琴,殷颖,等.基于自适应编辑距离的颜料光谱匹配识别方法[J].激光与光电子学进展,2018,55(11):113004.
[9] Zhang W T,Li Y W,Zhan P P,et al.Recognition of transgenic soybean oil based on terahertz time- domain spectroscopy and PCA-SVM[J].Infrared and Laser Engineering,2017,46(11):1125004.张文涛,李跃文,占平平,等.基于太赫兹时域光谱技术与PCA-SVM的转基因大豆油鉴别研究[J].红外与激光工程,2017,46(11):1125004.
[10] Pohl A,De?mann N,Dutzi K,et al.Identification of unknown substances by terahertz spectroscopy and multivariate data analysis[J].Journal of Infrared,Millimeter,and Terahertz Waves,2016,37(2):175-188.