基于太赫兹光谱和超香肠神经元网络的转基因甜菜的无损鉴别
|
摘要
提出了一种基于太赫兹光谱和超香肠神经元网络的检测方法,该方法能实现对转基因甜菜的无损鉴别。以转基因甜菜及其亲本为研究对象,通过太赫兹光谱仪扫描获得其在0.1～1.5 THz波段范围内的光谱数据,利用主成分分析方法提取能反映样本97.28%光谱信息的前3个主成分。对每个品种,随机挑选50个样本作为训练集合,建立超香肠神经元鉴别模型,挑选35个样本作为第1测试集,其他非同类样本作为第2测试集对鉴别模型进行验证。实验结果表明,鉴别模型对第1测试集的平均正确识别率达97.86%,对第2测试集的平均正确拒识率达95.08%,该方法具能较好的对转基因甜菜进行鉴别且准确率高。
This paper proposes a detection method based on terahertz spectroscopy and hyper sausage neuron(HSN) to fast and non-destructively distinguish transgenic substance. The transgenic sugar beets and their parents were studied and its Spectral data in the range of 0.1 to 1.5 THz were obtained by scanning through terahertz spectrometer, principal component analysis(PCA) is applied to extract spectral data of transgenic sugar beets, the cumulate reliabilities of the first three components were more than 97.28%. 50 samples were randomly selected as the training set in each variety and the recognition models were established base on hyper sausage neuron. 35 samples were selected as the first testing set and all the other non-similar samples were selected as the second testing set to verify the feasibility and effectiveness of this model, as the 95.08% samples in the second set were correctly rejected, the average rate of correct recognition in first testing set was 97.86%. The experiment results show that the detection method is feasible to distinguish transgenic sugar beets based on terahertz spectroscopy and hyper sausage neuron.
This paper proposes a detection method based on terahertz spectroscopy and hyper sausage neuron(HSN) to fast and non-destructively distinguish transgenic substance. The transgenic sugar beets and their parents were studied and its Spectral data in the range of 0.1 to 1.5 THz were obtained by scanning through terahertz spectrometer, principal component analysis(PCA) is applied to extract spectral data of transgenic sugar beets, the cumulate reliabilities of the first three components were more than 97.28%. 50 samples were randomly selected as the training set in each variety and the recognition models were established base on hyper sausage neuron. 35 samples were selected as the first testing set and all the other non-similar samples were selected as the second testing set to verify the feasibility and effectiveness of this model, as the 95.08% samples in the second set were correctly rejected, the average rate of correct recognition in first testing set was 97.86%. The experiment results show that the detection method is feasible to distinguish transgenic sugar beets based on terahertz spectroscopy and hyper sausage neuron.
引文
[1] 刘盛刚.太赫兹科学技术的新发展[J].中国基础科学,2006,8(1):7-12.LIU SH G. Recent development of terahertz science and technology[J]. China Basic Science,2006,8(1):7-12.
[2] LEE J H, CHOUNG M G. Nondestructive determination of herbicide-resistant genetically modified soybean seeds using near-infrared reflectance spectroscopy[J]. Food Chemistry,2011,126(1):368-373.
[3] IVANIRA M, SPACINO S I. Chemometric discrimination of genetically modified Coffea arabica cultivars using spectroscopic and chromatographic fingerprints[J]. Talanta, 2013,107(3):416-422.
[4] LIU J, LI ZH, HU F, et al. Identification of transgenic organisms based on terahertz spectroscopy and hyper sausage neuron[J]. Journal of Applied Spectroscopy,2015,82(1):104-110.
[5] 杨振刚,刘劲松,王可嘉.连续太赫兹成像系统对多层蜂窝样件无损检测的实验研究[J].光电子·激光,2013,24(6):1158-1162.YANG ZH G,LIU J S,WANG K J.Experimental research on nondestructive inspection for multilayer cellular samples using continuous terahertz waves imaging system [J].Journal of Optoelectronics·Laser,2013,24(6):1158-1162.
[6] MILCAMPS A, RABE S, CADE R, et al. Validity assessment of the detection method of maize event bt10 through investigation of its molecular structure[J]. Journal of Agricultural and Food Chemistry,2009,57(8):3156-3163.
[7] ZHANG W W,XIAO H N,QING L Y. Enhanced water vapour barrier and grease resistance of paper bilayer-coated with chitosan and beeswax[J]. Cabohydrate Polymers,2014,101 (9):401-406.
[8] 刘建军.太赫兹时域光谱技术在转基因物质检测上的识别方法研究[D]. 西安:西安电子科技大学,2015.LIU J J.Study on identification methods in the detection of transgenic material based on terahertz time domain spectroscopy[D].Xi’an: Xidian University,2015.
[9] 涂闪,张文涛,熊显名,等.基于太赫兹时域光谱系统的转基因棉花种子主成分特性分析[J].光子学报,2015,44(4):1-6.TU SH,ZHANG W T,XIONG X M, et al. Principal componet analysis for transgenic cotton seeds based on terahertz time domain spectroscopy system[J].Acta Photonica Sinica,2015,44(4):1-6.
[10] 聂君扬,张文涛,熊显名,等.基于太赫兹时域光谱技术与PCA-BPN网络的转基因大豆鉴别[J].光子学报.2016,45(5):1-7.NIE J Y, ZHANG W T, XIONG X M,et al. Recognition of transgenic soybeans based on terahertz spectroscopy and PCA-BPN network[J]. Acta Photonica Sinica,2016,45(5):1-7.
[11] 王道明,鲁昌华,蒋薇薇,等. 基于粒子群算法的决策树SVM多分类方法研究[J].电子测量与仪器学报,2015,49(4):611-615.WANG D M,LU CH H, JIANG W W, et al. Study on PSO-based decision-tree SVM multi-class classification method[J].Journal of Electronic Measurement and Instrumentation,2015,49(4):611-615.
[12] 蔡小娜,黄大庄,沈佐锐,等.用于昆虫分类鉴定的人工神经网络方法研究:主成分分析与数学建模[J].生物数学学报,2013,28(1):23-33.CAI X N,HUANG D ZH,SHEN Z R,et al. Research on artificial neural network method used for insects classification and identification principal componet analysis and mathematical modeling[J].Journal of Biomathematics, 2013, 28 (1):23-33.
[13] 曹芳,吴迪,何勇,等.基于可见-近红外反射光谱技术的葡萄品种鉴别方法的研究[J],光学学报,2009,29(2):537-540.CAO F, WU D, HE Y,et al. Variety discrimination of grapes based on visible-near reflection infrared spectroscopy[J].Acta Optica Sinica,2009,29(2):537-540.
[14] 翟亚锋,苏谦,邬文锦,等.基于仿生模式识别和近红外光谱的转基因小麦快速鉴别方法.[J].光谱学与光谱分析,2010,30(4):924-928.ZHAI Y F, SU Q, WU W J,et al.Fast discrimination of varieties of transgene wheat based on biomimetic pattern recognition and near infrared spectra[J]. Spectroscopy and Spectral Analysis,2010,30(4):924-928.
[15] BURNETT A D, FAN W H, UPADHYA P C, et al. Broadband terahertz time-domain spectroscopy of drugs-of-abuse and the use of principal component analysis[J].Analyst,2009,134(8): 1658-1668.
[16] 王守觉,徐建,王宪保,等.基于仿生模式识别的多镜头人脸身份确认系统研究[J].电子学报,2003,31(1):1-3.WANG SH J,XU J,WANG X B, et al.Multi-camera human-face personal identification system based on the biomimetic pattern recognition[J].Acta Electronica Sinica, 2003,31(1):1-3.
[2] LEE J H, CHOUNG M G. Nondestructive determination of herbicide-resistant genetically modified soybean seeds using near-infrared reflectance spectroscopy[J]. Food Chemistry,2011,126(1):368-373.
[3] IVANIRA M, SPACINO S I. Chemometric discrimination of genetically modified Coffea arabica cultivars using spectroscopic and chromatographic fingerprints[J]. Talanta, 2013,107(3):416-422.
[4] LIU J, LI ZH, HU F, et al. Identification of transgenic organisms based on terahertz spectroscopy and hyper sausage neuron[J]. Journal of Applied Spectroscopy,2015,82(1):104-110.
[5] 杨振刚,刘劲松,王可嘉.连续太赫兹成像系统对多层蜂窝样件无损检测的实验研究[J].光电子·激光,2013,24(6):1158-1162.YANG ZH G,LIU J S,WANG K J.Experimental research on nondestructive inspection for multilayer cellular samples using continuous terahertz waves imaging system [J].Journal of Optoelectronics·Laser,2013,24(6):1158-1162.
[6] MILCAMPS A, RABE S, CADE R, et al. Validity assessment of the detection method of maize event bt10 through investigation of its molecular structure[J]. Journal of Agricultural and Food Chemistry,2009,57(8):3156-3163.
[7] ZHANG W W,XIAO H N,QING L Y. Enhanced water vapour barrier and grease resistance of paper bilayer-coated with chitosan and beeswax[J]. Cabohydrate Polymers,2014,101 (9):401-406.
[8] 刘建军.太赫兹时域光谱技术在转基因物质检测上的识别方法研究[D]. 西安:西安电子科技大学,2015.LIU J J.Study on identification methods in the detection of transgenic material based on terahertz time domain spectroscopy[D].Xi’an: Xidian University,2015.
[9] 涂闪,张文涛,熊显名,等.基于太赫兹时域光谱系统的转基因棉花种子主成分特性分析[J].光子学报,2015,44(4):1-6.TU SH,ZHANG W T,XIONG X M, et al. Principal componet analysis for transgenic cotton seeds based on terahertz time domain spectroscopy system[J].Acta Photonica Sinica,2015,44(4):1-6.
[10] 聂君扬,张文涛,熊显名,等.基于太赫兹时域光谱技术与PCA-BPN网络的转基因大豆鉴别[J].光子学报.2016,45(5):1-7.NIE J Y, ZHANG W T, XIONG X M,et al. Recognition of transgenic soybeans based on terahertz spectroscopy and PCA-BPN network[J]. Acta Photonica Sinica,2016,45(5):1-7.
[11] 王道明,鲁昌华,蒋薇薇,等. 基于粒子群算法的决策树SVM多分类方法研究[J].电子测量与仪器学报,2015,49(4):611-615.WANG D M,LU CH H, JIANG W W, et al. Study on PSO-based decision-tree SVM multi-class classification method[J].Journal of Electronic Measurement and Instrumentation,2015,49(4):611-615.
[12] 蔡小娜,黄大庄,沈佐锐,等.用于昆虫分类鉴定的人工神经网络方法研究:主成分分析与数学建模[J].生物数学学报,2013,28(1):23-33.CAI X N,HUANG D ZH,SHEN Z R,et al. Research on artificial neural network method used for insects classification and identification principal componet analysis and mathematical modeling[J].Journal of Biomathematics, 2013, 28 (1):23-33.
[13] 曹芳,吴迪,何勇,等.基于可见-近红外反射光谱技术的葡萄品种鉴别方法的研究[J],光学学报,2009,29(2):537-540.CAO F, WU D, HE Y,et al. Variety discrimination of grapes based on visible-near reflection infrared spectroscopy[J].Acta Optica Sinica,2009,29(2):537-540.
[14] 翟亚锋,苏谦,邬文锦,等.基于仿生模式识别和近红外光谱的转基因小麦快速鉴别方法.[J].光谱学与光谱分析,2010,30(4):924-928.ZHAI Y F, SU Q, WU W J,et al.Fast discrimination of varieties of transgene wheat based on biomimetic pattern recognition and near infrared spectra[J]. Spectroscopy and Spectral Analysis,2010,30(4):924-928.
[15] BURNETT A D, FAN W H, UPADHYA P C, et al. Broadband terahertz time-domain spectroscopy of drugs-of-abuse and the use of principal component analysis[J].Analyst,2009,134(8): 1658-1668.
[16] 王守觉,徐建,王宪保,等.基于仿生模式识别的多镜头人脸身份确认系统研究[J].电子学报,2003,31(1):1-3.WANG SH J,XU J,WANG X B, et al.Multi-camera human-face personal identification system based on the biomimetic pattern recognition[J].Acta Electronica Sinica, 2003,31(1):1-3.