Optical Instruments, Volume. 42, Issue 4, 7(2020)

Efficient determination of water content in potato leaves based on spectroscopy technology

Xufeng YU... Hongmei LI, Wei ZHUO and Jie FENG* |Show fewer author(s)
Author Affiliations
  • College of Physics and Electronic Information, Yunnan Normal University, Kunming 650000, China
  • show less
    Figures & Tables(11)
    Schematic of hyperspectral imaging system
    Reflectance spectra of potato leaves
    Comparison of prediction results by PLSR models with whole spectra
    Comparison of prediction results by BP neural network models with whole spectra
    Comparison of prediction results by PLSR models with extracted spectra
    Prediction results by BP neural network models with extracted spectra
    • Table 1.

      Descriptive statistics of moisture content in the potato leaves

      马铃薯叶片水分含量统计

      View table
      View in Article

      Table 1.

      Descriptive statistics of moisture content in the potato leaves

      马铃薯叶片水分含量统计

      数据集样本数最大值/%最小值/%平均值/%平均偏差/%
      总样本11092.3580.0086.822.579
      建模集7592.3580.0086.922.575
      预测集3591.3380.0986.592.573
    • Table 2.

      All-band data PLSR model prediction results with different methods

      不同方法的全波段数据PLSR模型预测结果

      View table
      View in Article

      Table 2.

      All-band data PLSR model prediction results with different methods

      不同方法的全波段数据PLSR模型预测结果

      处理方法建模集验证集预测集
      Rc2RMSECRcv2RMSECVRp2RMSEP
      无预处理0.89070.85130.86580.69110.84930.9989
      SG平滑0.89070.85150.86560.69160.84930.9988
      MSC0.89060.85130.86570.69120.84920.9988
      SNV0.90680.78620.87250.67350.83631.0412
    • Table 3.

      All-band data BP neural network mode prediction results with different methods

      不同方法的全波段数据BP神经网络模型预测结果

      View table
      View in Article

      Table 3.

      All-band data BP neural network mode prediction results with different methods

      不同方法的全波段数据BP神经网络模型预测结果

      处理方法预测集
      Rp2RMSEP
      无预处理0.96550.4776
      SG平滑0.97470.4095
      MSC0.97910.3723
      SNV0.85190.9902
    • Table 4.

      Prediction results of different methods feature band PLSR model

      不同方法特征波段数据PLSR模型预测结果

      View table
      View in Article

      Table 4.

      Prediction results of different methods feature band PLSR model

      不同方法特征波段数据PLSR模型预测结果

      处理方法建模集验证集预测集
      Rc2RMSECRcv2RMSECVRp2RMSEP
      无预处理0.87830.89820.84070.75290.84681.0070
      SG平滑0.87720.90240.85280.72380.84920.9991
      MSC0.82121.17880.80360.85750.77450.9899
      SNV0.81641.10350.76240.91960.75201.2813
    • Table 5.

      Prediction results of BP neural network models with different methods

      不同方法特征波段数据BP神经网络模型预测结果

      View table
      View in Article

      Table 5.

      Prediction results of BP neural network models with different methods

      不同方法特征波段数据BP神经网络模型预测结果

      处理方法预测集
      Rp2RMSEP
      无预处理0.91700.7411
      SG平滑0.96580.4759
      MSC0.83511.0498
      SNV0.76601.2446
    Tools

    Get Citation

    Copy Citation Text

    Xufeng YU, Hongmei LI, Wei ZHUO, Jie FENG. Efficient determination of water content in potato leaves based on spectroscopy technology[J]. Optical Instruments, 2020, 42(4): 7

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: APPLICATION TECHNOLOGY

    Received: Mar. 30, 2020

    Accepted: --

    Published Online: Jan. 6, 2021

    The Author Email: FENG Jie (fengjie_yunnan@126.com)

    DOI:10.3969/j.issn.1005-5630.2020.04.002

    Topics