Spectroscopy and Spectral Analysis, Volume. 42, Issue 10, 3226(2022)

Early Detection of Tomato Gray Mold Disease With Multi-Dimensional Random Forest Based on Hyperspectral Image

Rong-hua GAO1、*, Lu FENG1、1; 2; *;, Yue ZHANG3、3;, Ji-dong YUAN3、3;, Hua-rui WU1、1; 2;, and Jing-qiu GU1、1; 2;
Author Affiliations
  • 11. Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China
  • 33. School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China
  • show less
    Figures & Tables(12)
    Tomato leaves hyperspectral image acquisition diagram
    Comparisons of reflectance of different ROIs after inoculation(a): Tomato leaf on the 6th day after infection; (b): Different ROIs selection;(c): Comparisons of reflectivity of different ROIs; (d): Comparisons of reflectivity of leaf on the 1st and 9th day after infection
    RGB image of leaves
    Changes of spectral reflectance of samples observed for 7 consecutive data
    Original series (a) and interrelated series (b)
    Symbolic aggregate approximation (SAX) method
    Symbolic Fourier approximation (SFA) method
    Hyperspectral curve of diseased and healthy leaves for 7 consecutive observations(a): Consecutive 7-day reflectivity of infected leaf 1; (b): Consecutive 7-day reflectivity of infected leaf 2; (c): Consecutive 7-day reflectivity of healthy leaf 1; (b): Consecutive 7-day reflectivity of healthy leaf 2
    Recognition results of SDSS-SAX-SFA-MRF model
    Recognition results of MDSS-SAX-SFA-MRF model(a): SFA multidimensional spectrum; (b): SAX multidimensional spectrum; (c): SAX+SFA multidimensional spectrum
    Comparison results of MDSS-SAX-SFA-MRF model and SDSS-SAX-SFA-MRF model
    • Table 1. Parameters of symbolic methods and weighted random forest model

      View table
      View in Article

      Table 1. Parameters of symbolic methods and weighted random forest model

      模型参数参数选择(程序随机)
      决策树数量50
      符号化方法{SAX, SFA, SAX+SFA}
      字母表大小a{3, 4, 5}
      单词长度l{3, 4, 5, 6}
      滑动窗口w{20%, 30%, 40%, 50%, 60%}
    Tools

    Get Citation

    Copy Citation Text

    Rong-hua GAO, Lu FENG, Yue ZHANG, Ji-dong YUAN, Hua-rui WU, Jing-qiu GU. Early Detection of Tomato Gray Mold Disease With Multi-Dimensional Random Forest Based on Hyperspectral Image[J]. Spectroscopy and Spectral Analysis, 2022, 42(10): 3226

    Download Citation

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

    Category: Research Articles

    Received: Aug. 11, 2021

    Accepted: Mar. 28, 2022

    Published Online: Nov. 23, 2022

    The Author Email: Rong-hua GAO (gaorh@nercita.org.cn)

    DOI:10.3964/j.issn.1000-0593(2022)10-3226-09

    Topics