基于梯度提升树算法的夏玉米叶面积指数反演
|
摘要
为了快速、准确、大范围获取大田夏玉米的叶面积指数(Leaf area index,LAI),基于实地采集的夏玉米LAI和株高,结合同时期的无人机多光谱影像,选择与夏玉米LAI相关性较强的8种植被指数以及株高作为反演LAI的输入变量,采用梯度提升树(Gradient boosting decision tree,GBDT)算法建立植被指数及株高与叶面积指数之间的预测模型,并与支持向量机(Support vector machine,SVM)和随机森林(Random forest,RF)算法建立的模型进行预测精度对比。结果表明,GBDT算法在3个样本组中的LAI预测值与实测值R~2分别为0. 571 0、0. 755 8、0. 644 1,均高于SVM算法(0. 547 2、0. 679 1、0. 616 8)和RF算法(0. 550 5、0. 697 3、0. 629 5);对应的RMSE分别为0. 002 7、0. 001 5、0. 001 6,均低于SVM算法(0. 211 7、0. 152 3、0. 159 7)和RF算法(0. 244 7、0. 214 7、0. 208 0)。该研究为快速准确的大田夏玉米LAI遥感监测提供了技术和方法。
Aiming to obtain the leaf area index( LAI) of large-scale summer maize in a high-efficiency,non-destructive and large-scale manner,and provide a technical reference for remote sensing monitoring of summer maize growth. The research was based on the field-collected summer maize LAI and maize height,as well as combined with multi-spectral data of the same period,eight vegetation indexes and height with strong correlation with summer maize LAI were selected as the input variables of gradient boosting decision tree( GBDT) algorithm model for LAI inversion. The support vector machine( SVM)model and the random forest( RF) model were taken as the reference models,which were used to compare the accuracy of prediction. The results showed that the GBDT algorithm model prediction consequence were better than the other two models among the three sample groups. R~2 of prediction value and measured LAI values of the sample groups 1,2 and 3 were 0. 571 0,0. 755 8 and 0. 644 1,respectively,which were higher than those of the SVM models( 0. 547 2,0. 679 1,0. 616 8) and RF models( 0. 550 5,0. 697 3,0. 629 5),corresponding root mean square error( RMSE) values were0. 002 7,0. 001 5 and 0. 001 6,which were lower than those of the SVM model( 0. 211 7,0. 152 3 and0. 159 7) and RF model( 0. 244 7,0. 214 7 and 0. 208 0). The research result provided a technical method for fast and accurate monitoring of summer maize LAI remote sensing in the field.
Aiming to obtain the leaf area index( LAI) of large-scale summer maize in a high-efficiency,non-destructive and large-scale manner,and provide a technical reference for remote sensing monitoring of summer maize growth. The research was based on the field-collected summer maize LAI and maize height,as well as combined with multi-spectral data of the same period,eight vegetation indexes and height with strong correlation with summer maize LAI were selected as the input variables of gradient boosting decision tree( GBDT) algorithm model for LAI inversion. The support vector machine( SVM)model and the random forest( RF) model were taken as the reference models,which were used to compare the accuracy of prediction. The results showed that the GBDT algorithm model prediction consequence were better than the other two models among the three sample groups. R~2 of prediction value and measured LAI values of the sample groups 1,2 and 3 were 0. 571 0,0. 755 8 and 0. 644 1,respectively,which were higher than those of the SVM models( 0. 547 2,0. 679 1,0. 616 8) and RF models( 0. 550 5,0. 697 3,0. 629 5),corresponding root mean square error( RMSE) values were0. 002 7,0. 001 5 and 0. 001 6,which were lower than those of the SVM model( 0. 211 7,0. 152 3 and0. 159 7) and RF model( 0. 244 7,0. 214 7 and 0. 208 0). The research result provided a technical method for fast and accurate monitoring of summer maize LAI remote sensing in the field.
引文
[1]MAIRE G L,MARSDEN C,VERHOEF W,et al.Leaf area index estimation with MODIS reflectance time series and model inversion during full rotations of Eucalyptus plantations[J].Remote Sensing of Environment,2011,115(2):586-599.
[2]CHEN J M,BLACK T A.Defining leaf-area index for non-flat leaves[J].Agricultural&Forest Meteorology,2010,15(4):421-429.
[3]刘镕源,王纪华,杨贵军,等.冬小麦叶面积指数地面测量方法的比较[J].农业工程学报,2011,27(3):220-224.LIU Rongyuan,WANG Jihua,YANG Guijun,et al.Comparision of ground-based LAI measuring methods on winter wheat[J].Transactions of the CSAE,2011,27(3):220-224.(in Chinese)
[4]惠凤鸣,田庆久,金震宇,等.植被指数与叶面积指数关系研究及定量化分析[J].遥感信息,2003(2):10-13.HUI Fengming,TIAN Qingjiu,JING Zhenyu,et al.Reasearch and quantitative analysis of the correlation between vegetation index and leaf area index[J].Remote Sensing Information,2003(2):10-13.(in Chinese)
[5]邢丽玮,李小娟,李昂晟,等.基于高光谱与多光谱植被指数的洪河沼泽植被叶面积指数估算模型对比研究[J].湿地科学,2013,11(3):313-319.XING Liwei,LI Xiaojuan,LI Angsheng,et al.A comparative study on estimation model for leaf area index of vegetation in marshes in Honghe national nature reserve based on hyperspectral and multispectral vegetation indices[J].Wetland Science,2013,11(3):313-319.(in Chinese)
[6]贾玉秋,李冰,程永政,等.基于GF-1与Landsat-8多光谱遥感影像的玉米LAI反演比较[J].农业工程学报,2015,31(9):173-179.JIA Yuqiu,LI Bing,CHENG Yongzheng,et al.Comparison between GF-1 images and Landsat-8 images in monitoring maize LAI[J].Transactions of the CSAE,2015,31(9):173-179.(in Chinese)
[7]刘照言,马灵玲,唐伶俐.基于SAIL模型的多角度多光谱遥感叶面积指数反演[J].干旱区地理,2010,33(1):93-98.LIU Zhaoyan,MA Lingling,TANG Lingli.Inversion of LAI based on SAIL model with multi-angle and multi-spectral remote sensing data[J].Arid Land Geography,2010,33(1):93-98.(in Chinese)
[8]王茜,宁纪锋,曹宇翔,等.基于SIFT算法的无人机遥感图像拼接技术[J].吉林大学学报(信息科学版),2017,35(2):188-197.WANG Qian,NING Jifeng,CAO Yuxiang,et al.Matching technologies of UAV remote sensing image based on SIFT[J].Journal of Jilin University(Information Science Edition),2017,35(2):188-197.(in Chinese)
[9]高林,杨贵军,于海洋,等.基于无人机高光谱遥感的冬小麦叶面积指数反演[J].农业工程学报,2016,32(22):113-120.GAO Lin,YANG Guijun,YU Haiyang,et al.Retrieving winter wheat leaf area index based on unmanned aerial vehicle hyperspectral remote sensing[J].Transactions of the CSAE,2016,32(22):113-120.(in Chinese)
[10]林卉,梁亮,张连蓬,等.基于支持向量机回归算法的小麦叶面积指数高光谱遥感反演[J].农业工程学报,2013,29(11):139-146.LIN Hui,LIANG Liang,ZHANG Lianpeng,et al.Wheat leaf area index inversion with hyperspectral remote sensing based on support vector regression algorithm[J].Transactions of the CSAE,2013,29(11):139-146.(in Chinese)
[11]李凤秀,张柏,宋开山,等.玉米叶面积指数与高光谱植被指数关系研究[J].遥感技术与应用,2007,22(5):586-592.LI Fengxiu,ZHANG Bai,SONG Kaishan,et al.Research and analysis of the correlation between hyperspectral vegetation index and leaf area index[J].Remote Sensing Technology&Application,2007,22(5):586-592.(in Chinese)
[12]袁媛,瑚波,武兴厚,等.夏玉米植被指数与叶面积指数的关系研究[J].陕西气象,2015(3):24-28.YUAN Yuan,HU Bo,WU Xinghou,et al.Research on the relationship between summer corn vegetation index and leaf area index[J].Journal of Shanxi Meteorology,2015(3):24-28.(in Chinese)
[13]梁栋,管青松,黄文江,等.基于支持向量机回归的冬小麦叶面积指数遥感反演[J].农业工程学报,2013,29(7):117-123.LIANG Dong,GUAN Qingsong,HUANG Wenjiang,et al.Remote sensing inversion of leaf area index based on support vector machine regression in winter wheat[J].Transactions of the CSAE,2013,29(7):117-123.(in Chinese)
[14]刘洋,刘荣高,陈镜明,等.叶面积指数遥感反演研究进展与展望[J].地球信息科学学报,2013,15(5):734-743.LIU Yang,LIU Ronggao,CHEN Jingming,et al.Current status and perspectives of leaf area index retrieval from optical remote sensing data[J].Journal of Geo-information Science,2013,15(5):734-743.(in Chinese)
[15]DURBHA S S,KING R L,YOUNAN N H.Support vector machines regression for retrieval of leaf area index from multiangle imaging spectroradiometer[J].Remote Sensing of Environment,2007,107(1-2):348-361.
[16]夏天,吴文斌,周清波,等.冬小麦叶面积指数高光谱遥感反演方法对比[J].农业工程学报,2013,29(3):139-147.XIA Tian,WU Wenbin,ZHOU Qingbo,et al.Comparison of two inversion methods for winter wheat leaf area index based on hyperspectral remote sensing[J].Transactions of the CSAE,2013,29(3):139-147.(in Chinese)
[17]谢巧云,黄文江,蔡淑红,等.冬小麦叶面积指数遥感反演方法比较研究[J].光谱学与光谱分析,2014,34(5):1352-1356.XIE Qiaoyun,HUANG Wenjiang,CAI Shuhong,et al.Comparative study on remote sensing invertion methods for estimating winter wheat leaf area index[J].Spectroscopy and Spectral Analysis,2014,34(5):1352-1356.(in Chinese)
[18]王丽爱,马昌,周旭东,等.基于随机森林回归算法的小麦叶片SPAD值遥感估算[J/OL].农业机械学报,2015,46(1):259-265.WANG Liai,MA Chang,ZHOU Xudong,et al.Estimation of wheat leaf SPAD value using RF algorithmic model and remote sensing data[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2015,46(1):259-265.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20150136&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2015.01.036.(in Chinese)
[19]张春兰,杨贵军,李贺丽,等.基于随机森林算法的冬小麦叶面积指数遥感反演研究[J].中国农业科学,2018,51(5):855-867.ZHANG Chunlan,YANG Guijun,LI Heli,et al.Remote sensing inversion of leaf area index of winter wheat based on random forest algorithm[J].Scientia Agricultura Sinica,2018,51(5):855-867.(in Chinese)
[20]HOUBORG R,MCCABE M F.A hybrid training approach for leaf area index estimation via Cubist and random forests machine-learning[J].ISPRS Journal of Photogrammetry&Remote Sensing,2017,135:173-188.
[21]岳继博,杨贵军,冯海宽.基于随机森林算法的冬小麦生物量遥感估算模型对比[J].农业工程学报,2016,32(18):175-182.YUE Jibo,YANG Guijun,FENG Haikuan.Comparative of remote sensing estimation models of winter wheat biomass based on random forest algorithm[J].Transactions of the CSAE,2016,32(18):175-182.(in Chinese)
[22]贾洁琼,刘万青.基于GF-1WFV影像和机器学习算法的玉米叶面积指数估算[J].中国图象图形学报,2018,23(5):107-117.JIA Jieqiong,LIU Wanqing.Estimation of maize leaf area index based on GF-1 WFV image and machine learning random algorithm[J].Journal of Image&Graphics,2018,23(5):107-117.(in Chinese)
[23]FRIEDMAN J H.Greedy function approximation:a gradient boosting machine[J].Annals of Statistics,2001,29(5):1189-1232.
[24]刘伟东,项月琴,郑兰芬,等.高光谱数据与水稻叶面积指数及叶绿素密度的相关分析[J].遥感学报,2000,4(4):279-283.LIU Weidong,XIANG Yueqin,ZHENG Lanfen,et al.Relationships between rice LAI,CH.D and hyperspectra data[J].Journal of Remote Sensing,2000,4(4):279-283.(in Chinese)
[25]VASUDEVAN A,KUMAR D A,BHUVANESWARI N S.Precision farming using unmanned aerial and ground vehicles[C]∥Technological Innovations in ICT for Agriculture and Rural Development.IEEE,2016.
[26]HAN J,SONG X,ZHOU Z,et al.Estimating leaf area index of winter oilseed rape using high spatial resolution satellite data[C]∥Fifth International Conference on Agro-geoinformatics.IEEE,2016.
[27]CHANG C C,LIN C J.LIBSVM:a library for support vector machines[J].Neural Computation,2011,13(9):2119-2147.
[28]CUTLER A,CUTLER D R,STEVENS J R.Random forests[J].Machine Learning,2004,45(1):157-176.
[29]FRIEDMAN J H.Stochastic gradient boosting[J].Computational Statistics&Data Analysis,2002,38(4):367-378.
[30]田嫦丽,张珣,潘博,等.互联网广告点击率预估模型中特征提取方法的研究与实现[J].计算机应用研究,2017,34(2):334-338.TIAN Changli,ZHANG Xun,PAN Bo,et al.Research and implementation of feature extraction methods on internet CTRprediction model[J].Application Research of Computers,2017,34(2):334-338.(in Chinese)
[2]CHEN J M,BLACK T A.Defining leaf-area index for non-flat leaves[J].Agricultural&Forest Meteorology,2010,15(4):421-429.
[3]刘镕源,王纪华,杨贵军,等.冬小麦叶面积指数地面测量方法的比较[J].农业工程学报,2011,27(3):220-224.LIU Rongyuan,WANG Jihua,YANG Guijun,et al.Comparision of ground-based LAI measuring methods on winter wheat[J].Transactions of the CSAE,2011,27(3):220-224.(in Chinese)
[4]惠凤鸣,田庆久,金震宇,等.植被指数与叶面积指数关系研究及定量化分析[J].遥感信息,2003(2):10-13.HUI Fengming,TIAN Qingjiu,JING Zhenyu,et al.Reasearch and quantitative analysis of the correlation between vegetation index and leaf area index[J].Remote Sensing Information,2003(2):10-13.(in Chinese)
[5]邢丽玮,李小娟,李昂晟,等.基于高光谱与多光谱植被指数的洪河沼泽植被叶面积指数估算模型对比研究[J].湿地科学,2013,11(3):313-319.XING Liwei,LI Xiaojuan,LI Angsheng,et al.A comparative study on estimation model for leaf area index of vegetation in marshes in Honghe national nature reserve based on hyperspectral and multispectral vegetation indices[J].Wetland Science,2013,11(3):313-319.(in Chinese)
[6]贾玉秋,李冰,程永政,等.基于GF-1与Landsat-8多光谱遥感影像的玉米LAI反演比较[J].农业工程学报,2015,31(9):173-179.JIA Yuqiu,LI Bing,CHENG Yongzheng,et al.Comparison between GF-1 images and Landsat-8 images in monitoring maize LAI[J].Transactions of the CSAE,2015,31(9):173-179.(in Chinese)
[7]刘照言,马灵玲,唐伶俐.基于SAIL模型的多角度多光谱遥感叶面积指数反演[J].干旱区地理,2010,33(1):93-98.LIU Zhaoyan,MA Lingling,TANG Lingli.Inversion of LAI based on SAIL model with multi-angle and multi-spectral remote sensing data[J].Arid Land Geography,2010,33(1):93-98.(in Chinese)
[8]王茜,宁纪锋,曹宇翔,等.基于SIFT算法的无人机遥感图像拼接技术[J].吉林大学学报(信息科学版),2017,35(2):188-197.WANG Qian,NING Jifeng,CAO Yuxiang,et al.Matching technologies of UAV remote sensing image based on SIFT[J].Journal of Jilin University(Information Science Edition),2017,35(2):188-197.(in Chinese)
[9]高林,杨贵军,于海洋,等.基于无人机高光谱遥感的冬小麦叶面积指数反演[J].农业工程学报,2016,32(22):113-120.GAO Lin,YANG Guijun,YU Haiyang,et al.Retrieving winter wheat leaf area index based on unmanned aerial vehicle hyperspectral remote sensing[J].Transactions of the CSAE,2016,32(22):113-120.(in Chinese)
[10]林卉,梁亮,张连蓬,等.基于支持向量机回归算法的小麦叶面积指数高光谱遥感反演[J].农业工程学报,2013,29(11):139-146.LIN Hui,LIANG Liang,ZHANG Lianpeng,et al.Wheat leaf area index inversion with hyperspectral remote sensing based on support vector regression algorithm[J].Transactions of the CSAE,2013,29(11):139-146.(in Chinese)
[11]李凤秀,张柏,宋开山,等.玉米叶面积指数与高光谱植被指数关系研究[J].遥感技术与应用,2007,22(5):586-592.LI Fengxiu,ZHANG Bai,SONG Kaishan,et al.Research and analysis of the correlation between hyperspectral vegetation index and leaf area index[J].Remote Sensing Technology&Application,2007,22(5):586-592.(in Chinese)
[12]袁媛,瑚波,武兴厚,等.夏玉米植被指数与叶面积指数的关系研究[J].陕西气象,2015(3):24-28.YUAN Yuan,HU Bo,WU Xinghou,et al.Research on the relationship between summer corn vegetation index and leaf area index[J].Journal of Shanxi Meteorology,2015(3):24-28.(in Chinese)
[13]梁栋,管青松,黄文江,等.基于支持向量机回归的冬小麦叶面积指数遥感反演[J].农业工程学报,2013,29(7):117-123.LIANG Dong,GUAN Qingsong,HUANG Wenjiang,et al.Remote sensing inversion of leaf area index based on support vector machine regression in winter wheat[J].Transactions of the CSAE,2013,29(7):117-123.(in Chinese)
[14]刘洋,刘荣高,陈镜明,等.叶面积指数遥感反演研究进展与展望[J].地球信息科学学报,2013,15(5):734-743.LIU Yang,LIU Ronggao,CHEN Jingming,et al.Current status and perspectives of leaf area index retrieval from optical remote sensing data[J].Journal of Geo-information Science,2013,15(5):734-743.(in Chinese)
[15]DURBHA S S,KING R L,YOUNAN N H.Support vector machines regression for retrieval of leaf area index from multiangle imaging spectroradiometer[J].Remote Sensing of Environment,2007,107(1-2):348-361.
[16]夏天,吴文斌,周清波,等.冬小麦叶面积指数高光谱遥感反演方法对比[J].农业工程学报,2013,29(3):139-147.XIA Tian,WU Wenbin,ZHOU Qingbo,et al.Comparison of two inversion methods for winter wheat leaf area index based on hyperspectral remote sensing[J].Transactions of the CSAE,2013,29(3):139-147.(in Chinese)
[17]谢巧云,黄文江,蔡淑红,等.冬小麦叶面积指数遥感反演方法比较研究[J].光谱学与光谱分析,2014,34(5):1352-1356.XIE Qiaoyun,HUANG Wenjiang,CAI Shuhong,et al.Comparative study on remote sensing invertion methods for estimating winter wheat leaf area index[J].Spectroscopy and Spectral Analysis,2014,34(5):1352-1356.(in Chinese)
[18]王丽爱,马昌,周旭东,等.基于随机森林回归算法的小麦叶片SPAD值遥感估算[J/OL].农业机械学报,2015,46(1):259-265.WANG Liai,MA Chang,ZHOU Xudong,et al.Estimation of wheat leaf SPAD value using RF algorithmic model and remote sensing data[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2015,46(1):259-265.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20150136&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2015.01.036.(in Chinese)
[19]张春兰,杨贵军,李贺丽,等.基于随机森林算法的冬小麦叶面积指数遥感反演研究[J].中国农业科学,2018,51(5):855-867.ZHANG Chunlan,YANG Guijun,LI Heli,et al.Remote sensing inversion of leaf area index of winter wheat based on random forest algorithm[J].Scientia Agricultura Sinica,2018,51(5):855-867.(in Chinese)
[20]HOUBORG R,MCCABE M F.A hybrid training approach for leaf area index estimation via Cubist and random forests machine-learning[J].ISPRS Journal of Photogrammetry&Remote Sensing,2017,135:173-188.
[21]岳继博,杨贵军,冯海宽.基于随机森林算法的冬小麦生物量遥感估算模型对比[J].农业工程学报,2016,32(18):175-182.YUE Jibo,YANG Guijun,FENG Haikuan.Comparative of remote sensing estimation models of winter wheat biomass based on random forest algorithm[J].Transactions of the CSAE,2016,32(18):175-182.(in Chinese)
[22]贾洁琼,刘万青.基于GF-1WFV影像和机器学习算法的玉米叶面积指数估算[J].中国图象图形学报,2018,23(5):107-117.JIA Jieqiong,LIU Wanqing.Estimation of maize leaf area index based on GF-1 WFV image and machine learning random algorithm[J].Journal of Image&Graphics,2018,23(5):107-117.(in Chinese)
[23]FRIEDMAN J H.Greedy function approximation:a gradient boosting machine[J].Annals of Statistics,2001,29(5):1189-1232.
[24]刘伟东,项月琴,郑兰芬,等.高光谱数据与水稻叶面积指数及叶绿素密度的相关分析[J].遥感学报,2000,4(4):279-283.LIU Weidong,XIANG Yueqin,ZHENG Lanfen,et al.Relationships between rice LAI,CH.D and hyperspectra data[J].Journal of Remote Sensing,2000,4(4):279-283.(in Chinese)
[25]VASUDEVAN A,KUMAR D A,BHUVANESWARI N S.Precision farming using unmanned aerial and ground vehicles[C]∥Technological Innovations in ICT for Agriculture and Rural Development.IEEE,2016.
[26]HAN J,SONG X,ZHOU Z,et al.Estimating leaf area index of winter oilseed rape using high spatial resolution satellite data[C]∥Fifth International Conference on Agro-geoinformatics.IEEE,2016.
[27]CHANG C C,LIN C J.LIBSVM:a library for support vector machines[J].Neural Computation,2011,13(9):2119-2147.
[28]CUTLER A,CUTLER D R,STEVENS J R.Random forests[J].Machine Learning,2004,45(1):157-176.
[29]FRIEDMAN J H.Stochastic gradient boosting[J].Computational Statistics&Data Analysis,2002,38(4):367-378.
[30]田嫦丽,张珣,潘博,等.互联网广告点击率预估模型中特征提取方法的研究与实现[J].计算机应用研究,2017,34(2):334-338.TIAN Changli,ZHANG Xun,PAN Bo,et al.Research and implementation of feature extraction methods on internet CTRprediction model[J].Application Research of Computers,2017,34(2):334-338.(in Chinese)