• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 8, 2397 (2022)
Ye-lan WU1,*, Hui-ning GUAN1,1;, Xiao-qin LIAN1,1;, Chong-chong YU1,1;..., Yu LIAO2,2; and Chao GAO1,1;|Show fewer author(s)
Author Affiliations
  • 11. Key Laboratory of Internet and Big Data in Light Industry, Beijing Technology and Business University, Beijing 100048, China
  • 22. Institute of Agricultural Engineering, Jiangxi Academy of Agricultural Sciences, Nanchang 330200, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2022)08-2397-06 Cite this Article
    Ye-lan WU, Hui-ning GUAN, Xiao-qin LIAN, Chong-chong YU, Yu LIAO, Chao GAO. Study on Detection Method of Leaves With Various Citrus Pests and Diseases by Hyperspectral Imaging[J]. Spectroscopy and Spectral Analysis, 2022, 42(8): 2397 Copy Citation Text show less
    All spectral curves of six types of leaf samples
    Fig. 1. All spectral curves of six types of leaf samples
    Average spectra of six types of leaf samples
    Fig. 2. Average spectra of six types of leaf samples
    Process of selecting characteristic wavelengths by CARS for original spectrum
    Fig. 3. Process of selecting characteristic wavelengths by CARS for original spectrum
    Loading curves of the first four principal components of the original spectrum
    Fig. 4. Loading curves of the first four principal components of the original spectrum
    Feature importance screening by XGBoost
    Fig. 5. Feature importance screening by XGBoost
    Modeling results of XGBoost using full-band
    Fig. 6. Modeling results of XGBoost using full-band
    Modeling results of SVM using Characteristic wavelengths
    Fig. 7. Modeling results of SVM using Characteristic wavelengths
    预处理方法CARS特征波长(λ/nm)PCA特征波长(λ/nm)
    Origin529, 534, 539, 712, 717, 749, 754, 760, 781, 787520, 554, 707, 712, 725, 731, 757
    1st Der529, 539, 754, 760, 792502, 689, 707, 712, 731, 741, 757
    MSC529, 534, 612, 701, 717, 733, 749, 754, 760, 781, 787, 792551, 678, 707, 712, 725, 731, 757
    MF529, 534, 622, 717, 754, 760, 776, 781, 787, 792529, 622, 707, 712, 725, 731, 757
    Table 1. Extracting characteristic wavelengths by CARS and PCA
    模型波长
    范围
    预处理评价指标
    OA/%Recall/%Train-time/s
    XGBoostFSOrigin94.3294.7759
    1stDer93.6094.0754
    MSC95.9894.3257
    MF96.5695.9163
    SVMCARSOrigin93.6394.03167
    1stDer90.2688.31182
    MSC87.9086.92184
    MF91.9590.74175
    PCAOrigin87.5391.00153
    1stDer90.8293.14162
    MSC83.5087.71159
    MF90.9890.87170
    Table 2. Modeling results of XGBoost and SVM
    Ye-lan WU, Hui-ning GUAN, Xiao-qin LIAN, Chong-chong YU, Yu LIAO, Chao GAO. Study on Detection Method of Leaves With Various Citrus Pests and Diseases by Hyperspectral Imaging[J]. Spectroscopy and Spectral Analysis, 2022, 42(8): 2397
    Download Citation