Journal of Applied Optics, Volume. 44, Issue 2, 420(2023)

Automatic detection method of forest diseases and insect pests based on spectral images

Xiaoli ZHOU1,*... Lijun ZHOU2, Lita YI1 and Yu LIU2 |Show fewer author(s)
Author Affiliations
  • 1Forest and Grassland Protection and Development Center of Ar Horqin Banner, Ar Horqin Banner 025550, China
  • 2Xi'an Institute of Applied Optics, Xi'an 710065, China
  • show less
    Figures & Tables(11)
    Processing flow of forest pests detection based on spectral images
    Comparison of relative spectral reflectance curves between healthy trees and infected trees
    Deep learning detection network structure
    Internal structure of pyramid pooling modules
    Internal structure of attention adjustment modules
    Test software interface
    Detection results of close infected area
    Detection results of remote infected area
    • Table 1. Comparison of detection rates in different scale disease areas

      View table
      View in Article

      Table 1. Comparison of detection rates in different scale disease areas

      区域大小/像素检出率/%
      大于等于50×5092.3
      大于30×30小于50×5065.6
      小于等于30×3023.1
    • Table 2. Comparison between proposed method and mainstream method

      View table
      View in Article

      Table 2. Comparison between proposed method and mainstream method

      模型图像分辨率/像素MIoU/%帧/s
      ENet [20]1 024×51253.160.5
      ESPNet [21]1 024×51256.5108.3
      BiSeNetV2[22]1 024×51265.5152.1
      BiSeNetV2-L[22]1 024×51268.9117
      STDC1-Seg50[23]1 024×51271.1198.6
      STDC2-Seg50[23]1 024×51272.3149.8
      本文方法1 024×51275.2137.1
    • Table 3. Results of ablation experiments

      View table
      View in Article

      Table 3. Results of ablation experiments

      金字塔池化模块注意力调整模块MIoU/%帧/s
      70.3156.6
      74.5113.8
      75.2137.1
    Tools

    Get Citation

    Copy Citation Text

    Xiaoli ZHOU, Lijun ZHOU, Lita YI, Yu LIU. Automatic detection method of forest diseases and insect pests based on spectral images[J]. Journal of Applied Optics, 2023, 44(2): 420

    Download Citation

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

    Category: Research Articles

    Received: May. 10, 2022

    Accepted: --

    Published Online: Apr. 14, 2023

    The Author Email:

    DOI:10.5768/JAO202344.0203009

    Topics