琯溪蜜柚叶片氮素含量多种高光谱估算模型对比研究

被引:11
作者
栗方亮 [1 ]
孔庆波 [1 ]
张青 [1 ]
庄木来 [2 ,3 ]
机构
[1] 福建省农业科学院土壤肥料研究所
[2] 福建省平和县农业农村局
[3] 平和琯溪蜜柚综合试验站
基金
国家重点研发计划;
关键词
蜜柚; 高光谱; 氮素; 光谱指数;
D O I
10.13925/j.cnki.gsxb.20210517
中图分类号
S666.3 [柚(文旦)]; S127 [遥感技术在农业上的应用];
学科分类号
082804 ;
摘要
【目的】蜜柚叶片氮素(nitrogen,N)含量是准确诊断和定量评价生长状况的重要指标,建立合适的蜜柚叶片氮素含量高光谱估算模型,为实现快速、无损、精确的氮素含量估测提供依据。【方法】基于蜜柚叶片高光谱数据和氮素含量实测数据,建立了蜜柚叶片偏最小二乘回归模型(PLS)、BP神经网络回归模型(BPNN)、随机森林回归模型(RF)和支持向量机回归模型(SVM),并确定了蜜柚叶片氮素含量最佳估算模型。【结果】原始光谱和一阶微分光谱与蜜柚叶片氮素含量在可见光范围内有多波段相关性显著,并出现多个极值。原始光谱曲线敏感波长为569 nm和704 nm,一阶微分曲线的敏感波长为541、617、695、753 nm。与蜜柚叶片氮素含量相关性较显著的光谱参量是NDVI′695,753、RVI′695,753、DVI′617,695、R′617、DVI′541,617。建立的PLS、BPNN、RF和SVM 4种蜜柚叶片氮素含量估算模型的决定系数R2分别为0.75、0.80、0.83和0.81,均方根误差RMSE分别为1.16、1.08、0.97和1.02。验证模型的决定系数R2分别为0.79、0.84、0.85和0.82,均方根误差RMSE分别为1.11、0.94、0.87和0.99,其估算模型的精确程度为RF>SVM>BPNN>PLS。【结论】通过对琯溪蜜柚叶片氮素含量进行4种高光谱估算模型对比,随机森林估算模型精度稍高于PLS、BPNN和SVM估算模型。研究结果为光谱监测蜜柚叶片氮素含量提供了技术依据。
引用
收藏
页码:882 / 891
页数:10
相关论文
共 34 条
  • [1] Study on the Quantitative Relationship Among Canopy Hyperspectral Reflectance, Vegetation Index and Cotton Leaf Nitrogen Content
    Yin, Caixia
    Lin, Jiao
    Ma, Lulu
    Zhang, Ze
    Hou, Tongyu
    Zhang, Lifu
    Lv, Xin
    [J]. JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING, 2021, 49 (08) : 1787 - 1799
  • [2] Using hyperspectral plant traits linked to photosynthetic efficiency to assess N and P partition.[J].Watt Michael S.;Buddenbaum Henning;Leonardo Ellen Mae C.;Estarija Honey Jane C.;Bown Horacio E.;Gomez-Gallego Mireia;Hartley Robin;Massam Peter;Wright Liam;Zarco-Tejada Pablo J..ISPRS Journal of Photogrammetry and Remote Sensing.2020,
  • [3] A Machine Learning Framework to Predict Nutrient Content in Valencia-Orange Leaf Hyperspectral Measurements
    Osco, Lucas Prado
    Marques Ramos, Ana Paula
    Faita Pinheiro, Mayara Maezano
    Saito Moriya, Erika Akemi
    Imai, Nilton Nobuhiro
    Estrabis, Nayara
    Ianczyk, Felipe
    de Araujo, Fabio Fernando
    Liesenberg, Veraldo
    de Castro Jorge, Lucio Andre
    Li, Jonathan
    Ma, Lingfei
    Goncalves, Wesley Nunes
    Marcato Junior, Jose
    Creste, Jose Eduardo
    [J]. REMOTE SENSING, 2020, 12 (06)
  • [4] Estimation and mapping of nitrogen content in apple trees at leaf and canopy levels using hyperspectral imaging.[J].Xujun Ye;Shiori Abe;Shuhuai Zhang.Precision Agriculture: An International Journal on Advances in Precision Agriculture.2020, 2
  • [5] Prediction Results of Different Modeling Methods in Soil Nutrient Concentrations Based on Spectral Technology
    Li, X-Y
    Fan, P-P
    Liu, Y.
    Hou, G-L
    Wang, Q.
    Lv, M-R
    [J]. JOURNAL OF APPLIED SPECTROSCOPY, 2019, 86 (04) : 765 - 770
  • [6] Modeling Mid-Season Rice Nitrogen Uptake Using Multispectral Satellite Data
    Brinkhoff, James
    Dunn, Brian W.
    Robson, Andrew J.
    Dunn, Tina S.
    Dehaan, Remy L.
    [J]. REMOTE SENSING, 2019, 11 (15)
  • [7] Selection of the optimal bands of first-derivative fluorescence characteristics for leaf nitrogen concentration estimation
    Yang, Jian
    Cheng, Yinjia
    Du, Lin
    Gong, Wei
    Shi, Shuo
    Sun, Jia
    Chen, Biwu
    [J]. APPLIED OPTICS, 2019, 58 (21) : 5720 - 5727
  • [8] Evaluation of Leaf N Concentration in Winter Wheat Based on Discrete Wavelet Transform Analysis
    Li, Fenling
    Wang, Li
    Liu, Jing
    Wang, Yuna
    Chang, Qingrui
    [J]. REMOTE SENSING, 2019, 11 (11)
  • [9] Prediction Model of Nitrogen Content in Apple Leaves based on Ground Imaging Spectroscopy.[J].Baichao Li;Xicun Zhu;Ruiyang Yu;Xiaoyan Guo;Shujing Cao;Huansan Zhao.Remote Sensing Science.2018,
  • [10] Impact of tillage practices on nitrogen accumulation and translocation in wheat and soil nitrate-nitrogen leaching in drylands.[J].Hongguang Wang;Zengjiang Guo;Yu Shi;Yongli Zhang;Zhenwen Yu.Soil & Tillage Research.2015,