基于小波变换和BP神经网络的水稻冠层重金属含量反演
|
摘要
自然农田生态系统中,农作物的各种生化参数受重金属污染胁迫后虽表现异常,但其特征往往极为微弱,极不稳定。利用处理非稳定信号方法中常用的信号处理方法——小波分析法(Db-5),对水稻的光谱反射率数据进行处理,有效提取光谱信号中受重金属污染胁迫而潜藏的一些"突变"弱信息。利用Db-5小波基进行小波变换,从中选取具有异常光谱特征的奇异点,利用奇异点对应波段(716、745、766 nm)的光谱反射率构建反向传播(BP)神经网络模型,对水稻冠层4种重金属含量进行反演。将利用模型得到的预测值与实测值进行相关性分析,结果表明,基于BP神经网络的水稻冠层重金属含量反演模型对于实验区镉、铅、汞、砷4种重金属胁迫,具有良好的反演效果。
In the "natural farmland ecosystem ",although the biochemical parameters of crops are abnormal under the stress of heavy metal pollution,their characteristics are often very weak,with small changes and extreme unstability. Wavelet analysis,a common signal processing method in unstable signal processing,was used to process spectral reflectance data of crops( rice) and effectively extract weak information of "mutation"hidden in spectral signals under the stress of heavy metal pollution. Wavelet transform was carried out by using Db-5 wavelet basis,and singular points with abnormal spectral characteristics were selected. Back propagation neural network model( BPNN) was constructed by using spectral reflectance of corresponding bands of singular points( 716 nm,745 nm and 766 nm) to invert the contents of four heavy metals in rice canopy. Correlation analysis was conducted between the predicted and measured values of the model,and the results showed that the inversion model of heavy metal content in rice canopy based on BP neural network had a good inversion effect on the stress of cadmium,lead,mercury and arsenic in the experimental area.
In the "natural farmland ecosystem ",although the biochemical parameters of crops are abnormal under the stress of heavy metal pollution,their characteristics are often very weak,with small changes and extreme unstability. Wavelet analysis,a common signal processing method in unstable signal processing,was used to process spectral reflectance data of crops( rice) and effectively extract weak information of "mutation"hidden in spectral signals under the stress of heavy metal pollution. Wavelet transform was carried out by using Db-5 wavelet basis,and singular points with abnormal spectral characteristics were selected. Back propagation neural network model( BPNN) was constructed by using spectral reflectance of corresponding bands of singular points( 716 nm,745 nm and 766 nm) to invert the contents of four heavy metals in rice canopy. Correlation analysis was conducted between the predicted and measured values of the model,and the results showed that the inversion model of heavy metal content in rice canopy based on BP neural network had a good inversion effect on the stress of cadmium,lead,mercury and arsenic in the experimental area.
引文
[1]肖相芬,张经廷,周丽丽,等.中国水稻重金属镉与铅污染GAP栽培控制关键点分析[J].中国农学通报,2009,25(21):130-136.XIAO Xiangfen,ZHANG Jingting,ZHOU Lili,et al. Analysis on key points for controlling cadmium and lead contamination during rice production[J]. Chinese Agricultural Science Bulletin,2009,25(21):130-136.(in Chinese)
[2]尚二萍,许尔琪,张红旗,等.中国粮食主产区耕地土壤重金属时空变化与污染源分析[J].环境科学,2018,39(10):4670-4683.SHANG Erping,XU Erqi,ZHANG Hongqi,et al. Spatial-temporal trends and pollution source analysis for heavy metal contamination of cultivated soils in five major grain producing regions of China[J]. Environment Science,2018,39(10):4670-4683.(in Chinese)
[3] GEORGE J W. Accumulation of cadmium in crop plants and its consequences to human health[J]. Advances in Agronomy,1993,51:173-212.
[4]周铮鹏.水稻叶绿素遥感反演及其与土壤重金属的相关性研究[D].长沙:长沙理工大学,2015.ZHOU Zhengpeng. Remote sensing quantitative inversion of the chlorophyll of paddy and research on the correlation between chlorophyll and soil heavy-metal[D]. Changsha:Changsha University of Science&Technology,2015.(in Chinese)
[5]张秋霞,张合兵,张会娟,等.粮食主产区耕地土壤重金属高光谱综合反演模型[J/OL].农业机械学报,2017,48(3):148-155.ZHANG Qiuxia,ZHANG Hebing,ZHANG Huijuan,et al. Hybrid inversion model of heavy metals with hyperspectral reflectance in cultivated soils of main grain producing areas[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2017,48(3):148-155. http://www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20170319&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2017. 03. 019.(in Chinese)
[6] LI Xuqing,LIU Xiangnan,LIU Meiling,et al. A hyperspectral index sensitive to subtle changes in the canopy chlorophyll content under arsnic stress[J]. International Journal of Applied Earth Observation and Geoinformation,2015,36:41-53.
[7]熊文成,聂忆皇,王桥,等.土壤污染遥感研究进展及应用展望[J].环境与可持续发展,2017,42(6):51-54.XIONG Wencheng,NIE Yihuang,WANG Qiao,et al. Progress in soil pollution remote sensing and its application prospect[J].Environmental and Sustainable Development,2017,42(6):51-54.(in Chinese)
[8] KOOISTRA L,SALAS E A L,CLEVERS J,et al. Exploring field vegetation reflectance as an indicator of soil contamination in river floodplains[J]. Environmental Pollution,2004,127(2):281-290.
[9] REN H Y,ZHUANG D F,PAN J J,et al. Hyper-spectral remote sensing to monitor vegetation stress[J]. Journal of Soils and Sediments,2008,8(5):323-326.
[10] LIU Y L,CHEN H,WU G F,et al. Feasibility of estimating heavy metal concentrations in Phragmites australis using laboratory-based hyperspectral data—a case study along Le'an River,China[J]. International Journal of Applied Earth Observation and Geoinformation,2010,12(Supp. 2):166-170.
[11]邬登巍,吴昀昭,马宏瑞.植物污染胁迫遥感监测研究综述[J].遥感技术与应用,2009,24(2):238-245.WU Dengwei,WU Yunzhao,MA Hongrui. Review on remote sensing monitoring on contaminated plant[J]. Remote Sensing Technology and Application,2009,24(2):238-245.(in Chinese)
[12] LIU M L,LIU X N,DING W C,et al. Monitoring stress levels on rice with heavy metals pollution from hyperspectral reflectance data using wavelet-fractal analysis[J]. International Journal of Applied Earth Observation and Geoinformation,2011,13(2):246-255.
[13]雷鸣,秦普丰,铁柏清.湖南湘江流域重金属污染的现状与分析[J].农业环境与发展,2010,27(2):62-65.LEI Ming,QIN Pufeng,TIE Boqing. Current situation and analysis of heavy metal pollution in Xiangjiang river basin in Hunan Province[J]. Agro-Environment and Development,2010,27(2):62-65.(in Chinese)
[14]胡杨,许涛.湖南湘江流域重金属污染现状及建议[J].民营科技,2016(8):220.HU Yang,XU Tao. Status and suggestions of heavy metal pollution in the Xiangjiang river basin in Hunan Province[J]. NonGovernmental Science&Technology,2016(8):220.(in Chinese)
[15] DEVORE R A, LUCIER B J. Fast wavelet techniques for near-optimal image processing[C]∥IEEE Military,Communications,Conference. IEEE Communications Society,1992:1129-1135.
[16] DONOHO D L. Nonlinear wavelet method for recovery of signals densities,and spectra from indirect and noisy data[C]∥Proceeding of Symposia in Applied Mathematics,1993:47.
[17] SHANKAR B. Multi-resolution biological transient extraction applied to respiratory crackles[J]. Compute. Boil. Med.,1996,26(1):25-39.
[18]陈劲松,朱博勤,邵芸.基于小波变换的多波段遥感图像条带噪声的去除[J].遥感信息,2003(2):6-9.CHEN Jinsong,ZHU Boqin,SHAO Yun. Destriping multi-sensor imagery based on wavelet transform[J]. Remote Sensing Information,2003(2):6-9.(in Chinese)
[19]陈涛,李兵海.基于小波变换去除遥感图像非高斯条纹噪音方法探讨[J].地质与勘探,2008,44(1):94-96.CHEN Tao,LI Binghai. Removal of non-Gaussian stripe noise in images based on wavelet[J]. Geology and Prospecting,2008,44(1):94-96.(in Chinese)
[20]李明慧.基于小波理论和Bi Shrink算法的遥感图像去噪研究[D].乌鲁木齐:新疆大学,2012.LI Minghui. Research of remote sensing image denoising based on wavelet theory and Bi Shrink algorithm[D]. Urumchi:Xinjiang University,2012.(in Chinese)
[21] MAIREG L,FRANCOIS C,DUFRENE E. Towards universal broad leaf chlorophyll indices using prospects imulated database and hyperspectral reflectance measurements[J]. Remote Sensing of Environment,2004,89(1):1-28.
[22] SIMON H. Neural networks:a comprehensive foundation[M]. 2版.北京:清华大学出版社,2001.
[23]王平,刘湘南,黄方.受污染胁迫玉米叶绿素含量微小变化的高光谱反演模型[J].光谱学与光谱学分析,2010,30(1):197-201.WANG Ping,LIU Xiangnan,HUANG Fang. Retrieval model fot subtle variation of contamination stressed marize chlorophyll using hyperspectral data[J]. Spectroscopy and Spectral Analysis,2010,30(1):197-201.(in Chinese)
[24]张德丰. MATLAB神经网络应用设计[M].北京:机械工业出版社,2009.
[25] GAUTAM R,PANIGRAHI S,FRANZEN D. Neural network optimisation of remotely sensed maize leaf nitrogen with a genetic algorithm and linear programming using five performance parameters[J]. Biosystems Engineering,2006,95(3):359-370.
[26]姚付启,张振华,杨润亚,等.基于主成分分析和BP神经网络的法国梧桐叶绿素含量高光谱反演研究[J].测绘科学,2010,35(1):109-112.YAO Fuqi,ZHANG Zhenhua,YANG Runya,et al. Research on Platanus orientalis L chlorophyll concentration estimation with hyperspectal data based on BP-artificial neural network and princial component anaylsis[J]. Science of Surveying and Mapping,2010,35(1):109-112.(in Chinese)
[27] LIU Meiling,LIU Xiangnan,WU Ling,et al. Wavelet-based detection of crop zinc stress assessment using hyperspectral reflectance[J]. Computers&Geosciences,2011,37(9):1254-1263.
[28]刘美玲.融合多元环境因子的水稻重金属污染水平遥感评估模型[D].北京:中国地质大学(北京),2011.LIU Meiling. Remote sensing evaluation model of heavy metal pollution level in rice with multi-environmental factors[D].Beijing:China University of Geosciences(Beijing),2011.(in Chinese)
[29]陈思宁,刘新会,侯娟,等.重金属锌胁迫的白菜叶片光谱响应研究[J].光谱学与光谱分析,2007,27(9):1797-1801.CHEN Sining,LIU Xinhui,HOU Juan,et al. Spectral response of cabbage leaf under heavy metal zinc stress[J]. Spectroscopy and Spectral Analysis,2007,27(9):1797-1801.(in Chinese)
[30]任红艳,庄大方,潘剑君,等.铅污染水稻的冠层高光谱特征研究[J].地球信息科学学报,2008,10(3):314-319.REN Hongyan,ZHUANG Dafang,PAN Jianjun,et al. Study on canopy hyperspectral characteristics of lead-contaminated rice[J]. Journal of Geomatic Information Science,2008,10(3):314-319.(in Chinese)
[31]刘美玲,刘湘南,李婷,等.水稻锌污染胁迫的光谱奇异性分析[J].农业工程学报,2010,26(3):191-197.LIU Meiling,LIU Xiangnan,LI Ting,et al. Spectral singularity analysis of zinc pollution in rice[J]. Transactions of the CSAE,2010,26(3):191-197.(in Chinese)
[2]尚二萍,许尔琪,张红旗,等.中国粮食主产区耕地土壤重金属时空变化与污染源分析[J].环境科学,2018,39(10):4670-4683.SHANG Erping,XU Erqi,ZHANG Hongqi,et al. Spatial-temporal trends and pollution source analysis for heavy metal contamination of cultivated soils in five major grain producing regions of China[J]. Environment Science,2018,39(10):4670-4683.(in Chinese)
[3] GEORGE J W. Accumulation of cadmium in crop plants and its consequences to human health[J]. Advances in Agronomy,1993,51:173-212.
[4]周铮鹏.水稻叶绿素遥感反演及其与土壤重金属的相关性研究[D].长沙:长沙理工大学,2015.ZHOU Zhengpeng. Remote sensing quantitative inversion of the chlorophyll of paddy and research on the correlation between chlorophyll and soil heavy-metal[D]. Changsha:Changsha University of Science&Technology,2015.(in Chinese)
[5]张秋霞,张合兵,张会娟,等.粮食主产区耕地土壤重金属高光谱综合反演模型[J/OL].农业机械学报,2017,48(3):148-155.ZHANG Qiuxia,ZHANG Hebing,ZHANG Huijuan,et al. Hybrid inversion model of heavy metals with hyperspectral reflectance in cultivated soils of main grain producing areas[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2017,48(3):148-155. http://www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20170319&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2017. 03. 019.(in Chinese)
[6] LI Xuqing,LIU Xiangnan,LIU Meiling,et al. A hyperspectral index sensitive to subtle changes in the canopy chlorophyll content under arsnic stress[J]. International Journal of Applied Earth Observation and Geoinformation,2015,36:41-53.
[7]熊文成,聂忆皇,王桥,等.土壤污染遥感研究进展及应用展望[J].环境与可持续发展,2017,42(6):51-54.XIONG Wencheng,NIE Yihuang,WANG Qiao,et al. Progress in soil pollution remote sensing and its application prospect[J].Environmental and Sustainable Development,2017,42(6):51-54.(in Chinese)
[8] KOOISTRA L,SALAS E A L,CLEVERS J,et al. Exploring field vegetation reflectance as an indicator of soil contamination in river floodplains[J]. Environmental Pollution,2004,127(2):281-290.
[9] REN H Y,ZHUANG D F,PAN J J,et al. Hyper-spectral remote sensing to monitor vegetation stress[J]. Journal of Soils and Sediments,2008,8(5):323-326.
[10] LIU Y L,CHEN H,WU G F,et al. Feasibility of estimating heavy metal concentrations in Phragmites australis using laboratory-based hyperspectral data—a case study along Le'an River,China[J]. International Journal of Applied Earth Observation and Geoinformation,2010,12(Supp. 2):166-170.
[11]邬登巍,吴昀昭,马宏瑞.植物污染胁迫遥感监测研究综述[J].遥感技术与应用,2009,24(2):238-245.WU Dengwei,WU Yunzhao,MA Hongrui. Review on remote sensing monitoring on contaminated plant[J]. Remote Sensing Technology and Application,2009,24(2):238-245.(in Chinese)
[12] LIU M L,LIU X N,DING W C,et al. Monitoring stress levels on rice with heavy metals pollution from hyperspectral reflectance data using wavelet-fractal analysis[J]. International Journal of Applied Earth Observation and Geoinformation,2011,13(2):246-255.
[13]雷鸣,秦普丰,铁柏清.湖南湘江流域重金属污染的现状与分析[J].农业环境与发展,2010,27(2):62-65.LEI Ming,QIN Pufeng,TIE Boqing. Current situation and analysis of heavy metal pollution in Xiangjiang river basin in Hunan Province[J]. Agro-Environment and Development,2010,27(2):62-65.(in Chinese)
[14]胡杨,许涛.湖南湘江流域重金属污染现状及建议[J].民营科技,2016(8):220.HU Yang,XU Tao. Status and suggestions of heavy metal pollution in the Xiangjiang river basin in Hunan Province[J]. NonGovernmental Science&Technology,2016(8):220.(in Chinese)
[15] DEVORE R A, LUCIER B J. Fast wavelet techniques for near-optimal image processing[C]∥IEEE Military,Communications,Conference. IEEE Communications Society,1992:1129-1135.
[16] DONOHO D L. Nonlinear wavelet method for recovery of signals densities,and spectra from indirect and noisy data[C]∥Proceeding of Symposia in Applied Mathematics,1993:47.
[17] SHANKAR B. Multi-resolution biological transient extraction applied to respiratory crackles[J]. Compute. Boil. Med.,1996,26(1):25-39.
[18]陈劲松,朱博勤,邵芸.基于小波变换的多波段遥感图像条带噪声的去除[J].遥感信息,2003(2):6-9.CHEN Jinsong,ZHU Boqin,SHAO Yun. Destriping multi-sensor imagery based on wavelet transform[J]. Remote Sensing Information,2003(2):6-9.(in Chinese)
[19]陈涛,李兵海.基于小波变换去除遥感图像非高斯条纹噪音方法探讨[J].地质与勘探,2008,44(1):94-96.CHEN Tao,LI Binghai. Removal of non-Gaussian stripe noise in images based on wavelet[J]. Geology and Prospecting,2008,44(1):94-96.(in Chinese)
[20]李明慧.基于小波理论和Bi Shrink算法的遥感图像去噪研究[D].乌鲁木齐:新疆大学,2012.LI Minghui. Research of remote sensing image denoising based on wavelet theory and Bi Shrink algorithm[D]. Urumchi:Xinjiang University,2012.(in Chinese)
[21] MAIREG L,FRANCOIS C,DUFRENE E. Towards universal broad leaf chlorophyll indices using prospects imulated database and hyperspectral reflectance measurements[J]. Remote Sensing of Environment,2004,89(1):1-28.
[22] SIMON H. Neural networks:a comprehensive foundation[M]. 2版.北京:清华大学出版社,2001.
[23]王平,刘湘南,黄方.受污染胁迫玉米叶绿素含量微小变化的高光谱反演模型[J].光谱学与光谱学分析,2010,30(1):197-201.WANG Ping,LIU Xiangnan,HUANG Fang. Retrieval model fot subtle variation of contamination stressed marize chlorophyll using hyperspectral data[J]. Spectroscopy and Spectral Analysis,2010,30(1):197-201.(in Chinese)
[24]张德丰. MATLAB神经网络应用设计[M].北京:机械工业出版社,2009.
[25] GAUTAM R,PANIGRAHI S,FRANZEN D. Neural network optimisation of remotely sensed maize leaf nitrogen with a genetic algorithm and linear programming using five performance parameters[J]. Biosystems Engineering,2006,95(3):359-370.
[26]姚付启,张振华,杨润亚,等.基于主成分分析和BP神经网络的法国梧桐叶绿素含量高光谱反演研究[J].测绘科学,2010,35(1):109-112.YAO Fuqi,ZHANG Zhenhua,YANG Runya,et al. Research on Platanus orientalis L chlorophyll concentration estimation with hyperspectal data based on BP-artificial neural network and princial component anaylsis[J]. Science of Surveying and Mapping,2010,35(1):109-112.(in Chinese)
[27] LIU Meiling,LIU Xiangnan,WU Ling,et al. Wavelet-based detection of crop zinc stress assessment using hyperspectral reflectance[J]. Computers&Geosciences,2011,37(9):1254-1263.
[28]刘美玲.融合多元环境因子的水稻重金属污染水平遥感评估模型[D].北京:中国地质大学(北京),2011.LIU Meiling. Remote sensing evaluation model of heavy metal pollution level in rice with multi-environmental factors[D].Beijing:China University of Geosciences(Beijing),2011.(in Chinese)
[29]陈思宁,刘新会,侯娟,等.重金属锌胁迫的白菜叶片光谱响应研究[J].光谱学与光谱分析,2007,27(9):1797-1801.CHEN Sining,LIU Xinhui,HOU Juan,et al. Spectral response of cabbage leaf under heavy metal zinc stress[J]. Spectroscopy and Spectral Analysis,2007,27(9):1797-1801.(in Chinese)
[30]任红艳,庄大方,潘剑君,等.铅污染水稻的冠层高光谱特征研究[J].地球信息科学学报,2008,10(3):314-319.REN Hongyan,ZHUANG Dafang,PAN Jianjun,et al. Study on canopy hyperspectral characteristics of lead-contaminated rice[J]. Journal of Geomatic Information Science,2008,10(3):314-319.(in Chinese)
[31]刘美玲,刘湘南,李婷,等.水稻锌污染胁迫的光谱奇异性分析[J].农业工程学报,2010,26(3):191-197.LIU Meiling,LIU Xiangnan,LI Ting,et al. Spectral singularity analysis of zinc pollution in rice[J]. Transactions of the CSAE,2010,26(3):191-197.(in Chinese)