Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (2): 353-361    DOI: 10.11873/j.issn.1004-0323.2021.2.0353
农业遥感专栏     
苹果叶片氮含量高光谱反演方法对比
杨福芹1,2(),冯海宽2(),李振海2,潘洁晨1,谢瑞1
1.河南工程学院土木工程学院,河南 郑州 451191
2.国家农业信息化工程技术研究中心,北京 100097
Comparison of Hyperspectral Remote Sensing Inversion Methods for Apple Leaf Nitrogen Content
Fuqin Yang1,2(),Haikuan Feng2(),Zhenhai Li2,Jiechen Pan1,Rui Xie1
1.College of Civil Engineering,Henan Institute of Engineering,Zhengzhou 451191,China
2.National Engineering Research Center for Information Technology in Agriculture,Beijing 100097,China
 全文: PDF(1898 KB)   HTML
摘要:

快速、无损、及时地准确估算苹果叶片氮含量是保证苹果产量和质量的基础,利用高光谱技术对苹果叶片氮含量进行遥感反演可为合理施肥提供理论依据。利用2012年和2013年山东省肥城市潮泉镇下寨村不同生育期的苹果叶片氮含量和相应的叶片光谱数据进行分析和建模。首先分析了叶片氮含量与原始光谱、一阶微分及三边光谱指数之间的相关性,筛选出对叶片氮含量敏感的光谱指数;构建了对叶片氮含量敏感的光谱指数NDSI和RSI;最后利用筛选的敏感光谱指数及构建的NDSI和RSI光谱指数,结合灰色关联分析(GRA)-偏最小二乘(PLS)方法及袋外数据重要性(OOB)-随机森林(RF)方法对叶片氮含量进行反演。结果表明:①叶片氮含量与原始光谱、一阶微分光谱之间的敏感波段分别为553、711 、527、708 和559 nm;构建的对叶片氮含量敏感的光谱指数分别为NDSI(567,615)和RSI(554,615);叶片氮含量对三边光谱指数之间相关性最好的光谱指数是SDy。②建模和验证结果表明用OOB-RF建立的苹果叶片氮含量估算模型具有较好的精度和可靠性,可以用来指导果树变量施肥,为监测氮素营养状况提供一种新的方法。

关键词: 苹果叶片叶片氮含量灰色关联分析随机森林偏最小二乘法    
Abstract:

Estimating nitrogen content of apple leaves rapidly non-destructive and timely is the basis of ensuring apple yield and quality, and the inversion of leaf nitrogen content using hyperspectral technology can provide theoretical basis for reasonable fertilization. The spectral and corresponding leaf nitrogen content of apple leaves were analyzed and modeling in apple critical growing stage from 2012 to 2013 in Feicheng, Shandong Province. Based on the above data, the correlation between leaf nitrogen content and original spectrum, first order differential spectrum, three-sided spectral index was firstly analysed in order to select sensitive spectral index of leaf nitrogen content; Secondly, the spectral index NDSI and RSI was built which were sensitive to leaf nitrogen content; Finally, the prediction model of the apple leaf nitrogen content was established based on the way that was grey correlation analysis-partial least squares regression and out-of-bag data- random forest algorithm. The results showed: (1) The sensitive bands between leaf nitrogen content and original spectrum and first-order differential spectrum were 553, 711, 527, 708 and 559 nm; the spectral indices sensitive to leaf nitrogen content were NDSI(567,615)and RSI(554,615); the best correlation between leaf nitrogen content and the three-sided spectral index was Sdy. (2) The result showed that OOB-RF estimation model had better accuracy and reliability, which can guide fruit tree variable fertilization using leaf nitrogen content. This way achieved prediction of leaf nitrogen content between regional and annual levels, and had a wide range of potential applications.

Key words: Apple leaf    Leaf nitrogen content    Grey relational analysis    Random forest    Partial least squares
收稿日期: 2019-12-28 出版日期: 2021-05-24
ZTFLH:  TP79  
基金资助: 国家自然科学基金项目(41601346);河南省科技攻关计划项目(202102310333);河南工程学院博士基金项目(D2017008)
通讯作者: 冯海宽     E-mail: yangfuqin0202@163.com;fenghaikuan123@163.com
作者简介: 杨福芹(1979-),女,河南安阳人,讲师,主要从事农业定量遥感研究。E?mail:yangfuqin0202@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
杨福芹
冯海宽
李振海
潘洁晨
谢瑞

引用本文:

杨福芹,冯海宽,李振海,潘洁晨,谢瑞. 苹果叶片氮含量高光谱反演方法对比[J]. 遥感技术与应用, 2021, 36(2): 353-361.

Fuqin Yang,Haikuan Feng,Zhenhai Li,Jiechen Pan,Rui Xie. Comparison of Hyperspectral Remote Sensing Inversion Methods for Apple Leaf Nitrogen Content. Remote Sensing Technology and Application, 2021, 36(2): 353-361.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.2.0353        http://www.rsta.ac.cn/CN/Y2021/V36/I2/353

日期富士生育期嘎啦生育期富士样本量嘎啦样本量
2012-05-10春梢旺长期春梢旺长期4416
2012-07-03春梢停长期春梢停长期4315
2012-08-10果实膨大期果实成熟期4416
2012-09-20果实膨大期叶变色期4316
2012-10-18果实成熟期叶变色期4416
2013-04-20开花盛期开花盛期279
2013-05-25春梢旺长期春梢旺长期279
2013-07-18秋梢旺长期秋梢旺长期279
2013-08-30果实膨大期果实成熟期279
2013-10-25果实成熟期叶变色期279
总体353124
表1  试验测定数据的详细列表
类型光谱指数定义
基于光谱位置变量Dr红边内最大一阶微分值
λrDr对应波长
Db蓝边内最大一阶微分
λbDb对应波长
Dy黄边内最大一阶微分值
λyDy对应波长
Rg绿峰反射率
λgRg对应波长
Ro红谷反射率
基于光谱面积变量SDr红边内一阶微分总和
SDb蓝边内一阶微分总和
SDy黄边内一阶微分总和
基于光谱指数变量SDr/SDb

比值光谱指数

SDr/Sdy
Rg/Ro
(SDr-SDb)/(SDr+SDb)

归一化光谱指数

(SDr-SDy)/(SDr+SDy)
(Rg-Ro)/(Rg+Ro)
表2  光谱指数定义
图1  原始光谱与叶片全氮含量的相关性
光谱指数回归方程建模(n=180)验证(n=297)
R2RMSERE/%R2RMSERE/%
R553y=23.505x+0.650 50.580.3910.800.030.4513.89
R724y=15.766x-2.512 20.680.349.260.180.4011.68
表3  基于原始光谱指数的叶片氮含量预测模型及验证
图2  一阶微分光谱与叶片氮含量的相关性
光谱指数回归方程建模(n=180)验证(n=297)
R2RMSERE/%R2RMSERE/%
D527y=1.477ln(x)+12.440.750.308.150.380.3712.13
D559y=-2 473.9x+1.068 80.800.277.390.440.3410.78
D708y=440.7x-1.342 50.760.298.090.350.3912.74
表4  基于一阶微分参数的叶片氮含量含量预测模型及验证
图3  NDSI与RSI和LNC决定系数等高线图
光谱指数回归方程建模(n=180)验证(n=297)
R2RMSERE/%R2RMSERE/%
NDSI(567,615)y=12.042x+0.624 70.800.277.140.520.3612.11
RSI(554,615)y=2.8449x-1.841 10.810.267.050.570.3511.75
表5  基于NDSI和RSI的叶片氮含量预测模型及验证
光谱指数相关系数光谱指数相关系数
Dr0.19*SDr0.39**
λr-0.77**SDb0.83**
Db0.85**SDy-0.88**
λb0.22**SDr/SDb-0.84**
Dy0.83**SDr/SDy0.87**
λy-0.16*Rg/Ro0.85**
Rg0.76**(SDr-SDb)/(SDr+SDb)-0.82**
λg0.45**(SDr-SDy)/(SDr+SDy)0.87**
Ro-0.27**(Rg-Ro)/(Rg+Ro)0.87**
表6  叶片氮含量与三边光谱指数间的相关系数
光谱指数回归方程建模(n=180)验证(n=297)
R2RMSERE%R2RMSERE%
λry = -0.0962x + 72.0150.600.3810.440.460.3511.09
Dby = 1.704ln(x) + 13.6620.750.308.260.400.3510.89
Dyy = 2766.6x + 2.13530.680.349.190.320.3811.80
Rgy = 23.509x + 0.64950.580.3910.810.000.4513.88
SDby = 1.6986ln(x) + 8.67030.720.318.340.130.4414.36
SDyy = -58.941x + 1.32190.780.287.460.410.3611.70
SDr/SDby = -1.741ln(x) + 7.56220.710.328.550.200.4314.12
SDr/SDyy = 0.1131x + 5.12340.750.308.420.400.3913.06
Rg/Roy = 2.7425ln(x) + 0.94050.750.307.910.330.4113.71
(SDr-SDb)/(SDr+SDb)y = -11.617x + 12.9690.680.348.730.210.4213.79
(SDr-SDy)/(SDr+SDy)y = 14.562ln(x) + 1.30120.750.308.000.470.3511.56
(Rg-Ro)/(Rg+Ro)y = 6.521x + 0.68270.750.307.950.320.4113.92
表7  基于三边光谱指数的叶片氮含量预测模型及验证
光谱指数灰色关联排序光谱指数灰色关联排序
g/Ro0.9071SDr0.83014
NDSI(567,615)0.9052Dr0.82815
RSI(554,615)0.9003λb0.82416

(Rg-Ro)/

(Rg+Ro)

0.8984λg0.82317
D7080.8975λy0.82218
D5590.8816λr0.82019
Rg0.8797D5270.81320
R5530.8788

(SDr-SDb)/

(SDr+SDb)

0.80021
R7240.8739Ro0.78122
SDy0.85910Dy0.71723
SDb0.84611SDr/SDb0.71624

(SDr-SDy)/

(SDr+SDy)

0.84412SDr/SDy0.70825
Db0.84213
表8  光谱指数与叶片氮含量灰色关联分析顺序
光谱指数袋外数据排序光谱指数袋外数据排序
Db10.2041SDr/SDb1.49514
D5278.6962Rg/Ro1.15415
R5534.6823Dr1.00016
D5594.6774SDy0.98217
RSI(554,615)4.1585

(SDr-SDb)/

(SDr+SDb)

0.94718
Rg3.3276

(Rg-Ro)/

(Rg+Ro)

0.94619
D7083.2987SDr0.64020
Dy3.0688Ro0.47121
SDr/SDy3.0339λr0.45322

(SDr-SDy)/

(SDr+SDy)

2.97110λb0.30623
NDSI(567,615)2.82311λg0.10024
SDb2.43512λy0.013225
R7242.02213
表9  光谱指数与叶片氮含量袋外数据重要性顺序
回归模型建模(n=180)验证(n=297)
R2RMSERE/%R2RMSERE/%
GRA-PLS0.820.257.010.530.3511.53
OOB-RF0.860.235.760.540.309.18
表10  基于GRA-PLS和OOB-RF叶片氮含量预测模型及验证
图4  叶片氮含量预测值与实测值的关系
1 Delalieux S, Somers B, Verstraeten W W, et al. Hyperspectral Indices to Diagnose Leaf Biotic Stress of Apple Plants, Considering Leaf Phenology[J]. International Journal of Remote Sensing, 2009, 30(8):1887-1912.
2 Penuelas J, Filella I, Lloret P, et al. Reflectance Assessment of Mite Effects on Apple Trees[J]. International Journal of Remote Sensing, 1995, 16(14):2727-2733.
3 Tong Qingxi, Zhang Bing, Zheng Lanfen. Hyperspectral Remote Sensing: principle, Technology and Application[M].Beijing: Higher Education Press, 2006.童庆禧, 张兵, 郑兰芬.高光谱遥感—原理、技术与应用[M]. 北京: 高等教育出版社, 2006.
4 Xue L, Cao W, Luo W, et al. Monitoring Leaf Nitrogen Status in Rice with Canopy Spectral Reflectance[J]. Agronomy Journal, 2004, 96(1):135-142.
5 Zhu Y, Yao X, Tian Y C, et al. Analysis of Common Canopy Vegetation Indices for Indicating Leaf Nitrogen Accumulations in Wheat and Rice[J]. International Journal of Applied Earth Observation & Geoinformation, 2008, 10(1):1-10.
6 Haboudane D, Miller J R, Tremblay N, et al. Integrated Narrow-band Vegetation Indices for Prediction of Crop Chlorophyll Content for Application to Precision Agriculture[J]. Remote Sensing of Environment, 2002, 81(2):416-426.
7 Haboudane D, Miller J R, Pattey E, et al. Hyperspectral Vegetation Indices and Novel Algorithms for Predicting Green LAI of Crop Canopies: Modeling and Validation in the Context of Precision Agriculture[J]. Remote Sensing of Environment, 2004, 90(3):337-352.
8 Yang Fuqin, Feng Haikuan, Li Zhenhai, et al. Estimation of Leaf Area Index of Winter Wheat based on Akaike’s Information Criterion[J]. Transactions of the Chinese Society of Agricultural Machinery, 2015,46(11): 112-120.
8 杨福芹,冯海宽,李振海, 等. 基于赤池信息量准则的冬小麦叶面积指数估算[J]. 农业机械学报, 2015,46(11): 112-120.
9 Fu Yuanyuan, Wang Jihua, Yang Guijun, et al. Band Depth Analysis and Partial Least Square Regression based Winter Wheat Biomass Estimation Using Hyperspectral Measurements[J].Spectroscopy and Spectral Analysis, 2013, 33(5): 1315-1319.
9 付元元,王纪华,杨贵军,等. 应用波段深度分析和偏最小二乘回归的冬小麦生物量高光谱估算[J]. 光谱学与光谱分析, 2013, 33(5): 1315-1319.
10 Miphokasap P, Honda K, Vaiphasa C, et al. Estimating Canopy Nitrogen Concentration in Sugarcane Using Field Imaging Spectroscopy[J]. Remote Sensing,2012,4(6):1651-1670.
11 Stroppiana D, Boschetti M, Brivio P A, et al. Plant Nitrogen Concentration in Paddy Rice from Field Canopy Hyperspectral Radiometry[J].Field Crops Research,2009,111(1):119-129.
12 Yao X, Zhu Y, Tian Y C, et al. Exploring Hyperspectral Bands and Estimation Indices for Leaf Nitrogen Accumulation in Wheat[J]. International Journal of Applied Earth Observation and Geoinformation, 2010, 12(2): 89-100.
13 Li Bingzhi, Li Minxia, Zhou Xuan, et al. Hyperspectral Estimation Models for Nitrogen Contents of Apple Leaves[J]. Journal of Remote Sensing, 2010,14(4):761-773.
13 李丙智,李敏夏,周璇,等.苹果树叶片全氮含量高光谱估算模型研究[J]. 遥感学报,2010,14(4):761-773.
14 Zhu Xicun, Zhao Gengxing, Wang Ling, et al. Hyperspectrum based Prediction Model for Nitrogen Content of Apple Flowers[J]. Spectroscopy and Spectral Analysis, 2010, 30(2): 416-420.
14 朱西存, 赵庚星, 王凌, 等. 基于高光谱的苹果花氮素含量预测模型研究[J]. 光谱学与光谱分析, 2010, 30(2): 416-420.
15 Yang Fuqin, Feng Haikuan, Li Zhenhai, et al. Prediction Model for Nitrogen Content of Apple Leaves based on Spectral Features from Visible and Near Infrared Regions[J]. Transactions of the Chinese Society of Agricultural Machinery, 2017,48(9):143-151.
15 杨福芹,冯海宽,李振海, 等.基于可见光-近红外光谱特征参数的苹果叶片氮含量预测[J].农业机械学报, 2017, 48(9): 143-151.
16 Zhang G C, Li Z, Yan X M, et al. Rapid Analysis of Apple Leaf Nitrogen Using near Infrared Spectroscopy and Multiple Linear Regression[J]. Communications in Soil Science and Plant Analysis, 2012,43( 13) : 1768 -1772.
17 Yue Xuejun, Quan Dongping, Hong Tiansheng, et al. Estimation Model of Nitrogen Content for Citrus Leaves by Spectral Technology based on Manifold Learning Algorithm[J]. Transactions of the Chinese Society of Agricultural Machinery, 2015, 46(6):244-250.
17 岳学军, 全东平, 洪添胜, 等. 基于流形学习算法的柑橘叶片氮含量光谱估测模.[J].农业机械学报, 2015, 46(6): 244-250.
18 Chen Lan, Chang Qingrui, Gao Yifan, et al. Hyperspectral Estimationmodel of Chlorophyll Content in Kiwifruit Leaves[J]. Journal of Northwest A&F University(Natural Science Edition),2020, 48(6):1-12.
18 陈澜, 常庆瑞, 高一帆, 等. 猕猴桃叶片叶绿素含量高光谱估算模型研究[J].西北农林科技大学学报(自然科学版), 2020,48(6): 1-12.
19 Song Xiao, Xu Ruiyang, Huang Shaomin. et al. Nitrogen Content Inversion of Wheat Canopy Leaf based on Ground Spectral Reflectance Data[J]. Chinese Journal of Applied Ecology, 2020, 31(5):1636-1644.
19 宋晓,许瑞阳,黄绍敏, 等.基于地面观测光谱数据的冬小麦冠层叶片氮含量反演模型[J]. 应用生态学报, 2020, 31(5): 1636-1644.
20 Feng Haikuan, Yang Fuqin, Li Zhenhai, et al. Hyperspectral Estimation of Leaf Total Phosphorus Content in Apple Tree based on Optimal Weights Combination Model[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(7): 173-180.
20 冯海宽, 杨福芹, 李振海, 等. 最优权重组合模型和高光谱估算苹果叶片全磷含量[J]. 农业工程学报, 2016, 32(7): 173-180.
21 Feng Haikuan, Li Zhenhai, Jin Xiuliang, et al. Estimating Equivalent Water Thickness of Apple Leaves Using Hypersecptral Data based on EFAST and PLS[J].Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(12): 165-171.
21 冯海宽, 李振海, 金秀良,等. 基于EFAST和PLS的苹果叶片等效水厚度高光谱估算[J].农业工程学报, 2016, 32(12): 165-171.
22 Deng Julong. Course of Grey System Theory[M].Beijing: Huazhong University of Science and Technology Press, 1990.
22 邓聚龙.灰色系统理论教程[M]. 北京: 华中理工大学出版社, 1990
23 Breiman L. Out-Of-Bag Estimation[R]. Cites Feer, Technical Report, 1996.
24 Wang Guizeng. Principal Component Analysis and Partial Least Squares[M]. Beijing: Tsinghua University Press, 2012.
24 王桂增.主元分析与偏最小二乘法[M].北京:清华大学出版社, 2012.
25 Breiman L.Random F[J].Machine Learning,2001,45(1):5-32.
26 Wang Ling, Zhao Gengxing, Zhu Xicun, et al. Satellite Remote Sensing Retrieval of Canopy Nitrogen Nutritional Status of Apple Rrees at Blossom Stage[J]. Chinese Journal of Applied Ecology, 2013, 24(10): 2863-2870.
26 王凌, 赵庚星, 朱西存,等. 花期苹果树冠氮素营养状况的卫星遥感反演[J]. 应用生态学报, 2013, 24(10): 2863-2870.
27 Fang Kuangnan,Wu Jianbin,Zhu Jianping,et al. A Review of Technologies on Random Forests[J].Statistics & Information Forum,2011,26(3):32-38.
27 方匡南,吴见彬,朱建平,等.随机森林方法研究综述[J].统计与信息论坛,2011,26(3):32-38.
28 Cutler R D, Edwards T C, Beard K H, et al. Random Forests for Classification in Ecology[J]. Ecology, 2007, 88(11): 2783-2792.
29 Li Z,Nie C,Wei C,et al.Comparison of Four Chemometric Te-chniques for Estimating Leaf Nitrogen Concentrations in Winter Wheat(Triticum Aestivum) based on Hyperspectral Featu-res[J].Journal of Applied Spectroscopy,2016,83(2):240-247.
[1] 陈彦四,黄春林,侯金亮,韩伟孝,冯娅娅,李翔华,王静. 基于多时相Sentinel-2影像的黑河中游玉米种植面积提取研究[J]. 遥感技术与应用, 2021, 36(2): 324-331.
[2] 王崇阳,田昕. 基于GF⁃1 PMS数据的森林覆盖变化检测[J]. 遥感技术与应用, 2021, 36(1): 208-216.
[3] 付甜梦,张丽,陈博伟,闫敏. 基于GEE平台的海岛地表覆盖提取及变化监测—以苏拉威西岛为例[J]. 遥感技术与应用, 2021, 36(1): 55-64.
[4] 马腾耀,肖鹏峰,张学良,马威,郭金金. 基于特征优选的GF-3全极化数据积雪识别[J]. 遥感技术与应用, 2020, 35(6): 1292-1302.
[5] 王一帆,徐涵秋. 基于客观阈值与随机森林Gini指标的水体遥感指数对比[J]. 遥感技术与应用, 2020, 35(5): 1089-1098.
[6] 李萌,年雁云,边瑞,白艳萍,马金辉. 基于多源遥感影像的青海云杉和祁连圆柏分类[J]. 遥感技术与应用, 2020, 35(4): 855-863.
[7] 张坤,刘乃文,高帅,赵书慧. 数据驱动的植被总初级生产力估算方法研究[J]. 遥感技术与应用, 2020, 35(4): 943-949.
[8] 董超,赵庚星. 时序数据集构建质量对土地覆盖分类精度的影响研究[J]. 遥感技术与应用, 2020, 35(3): 558-566.
[9] 李净,温松楠. 基于3种机器学习法的太阳辐射模拟研究[J]. 遥感技术与应用, 2020, 35(3): 615-622.
[10] 唐廷元,付波霖,何素云,娄佩卿,闭璐. 基于GF-1和Sentinel-1A的漓江流域典型地物信息提取[J]. 遥感技术与应用, 2020, 35(2): 448-457.
[11] 柴旭荣,李明,周义,王金风,田庆春. 影像的土地覆被快速分类[J]. 遥感技术与应用, 2020, 35(2): 315-325.
[12] 刘培,余志远,马威,韩瑞梅,陈正超,王涵,杨磊库. 基于地形信息的Landsat与Radarsat-2遥感数据协同分类研究[J]. 遥感技术与应用, 2019, 34(6): 1269-1275.
[13] 李哲,张沁雨,彭道黎. 基于高分二号遥感影像的树种分类方法[J]. 遥感技术与应用, 2019, 34(5): 970-982.
[14] 程凯, 王卷乐, Jaahanaa Davaadorj, 韩雪华. 近30 a来蒙古国乌兰巴托市城镇扩张及其驱动力分析[J]. 遥感技术与应用, 2019, 34(1): 90-100.
[15] 谷晓天, 高小红, 马慧娟, 史飞飞, 刘雪梅, 曹晓敏. 复杂地形区土地利用/土地覆被分类机器学习方法比较研究[J]. 遥感技术与应用, 2019, 34(1): 57-67.