Spectroscopy and Spectral Analysis, Volume. 41, Issue 12, 3701(2021)

Vis-NIR Spectra Discriminant of Pesticide Residues on the Hami Melon Surface by GADF and Multi-Scale CNN

Guo-wei YU1、*, Ben-xue MA1、1; 2; *;, Jin-cheng CHEN1、1; 3;, Fu-min DANG4、4; 5;, Xiao-zhan LI1、1;, Cong LI1、1;, and Gang WANG1、1;
Author Affiliations
  • 11. College of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, China
  • 44. Food Quality Supervision and Testing Center (Shihezi), Ministry of Agriculture, Shihezi 832000, China
  • show less
    Figures & Tables(7)
    Vis-NIR spectra collection system
    The multi-scale CNN structure
    Vis-NIR spectra of pesticide residues on the Hami melon surface(a): Raw spectra; (b): Average spectra
    GAF image of Vis-NIR spectra(a1, a2, a3): No residues; (b1, b2, b3): Chlorothalonil residues; (c1, c2, c3): Imidacloprid residues;(1): Standardization; (2): GASF; (3): GADF
    The training results of the multi-scale CNN model(a): GASF; (b): GADF
    The confusion matrixof the multi-scale CNN model(a): GASF; (b): GADF
    • Table 1. Test performance comparisons of different classification models

      View table
      View in Article

      Table 1. Test performance comparisons of different classification models

      模型正确判别个数判别准确
      率/%
      无残留百菌清吡虫啉
      多尺度CNN20192098.33
      AlexNet20191793.33
      VGG-1620191895.00
      SVM19181790.00
      ELM19181891.67
    Tools

    Get Citation

    Copy Citation Text

    Guo-wei YU, Ben-xue MA, Jin-cheng CHEN, Fu-min DANG, Xiao-zhan LI, Cong LI, Gang WANG. Vis-NIR Spectra Discriminant of Pesticide Residues on the Hami Melon Surface by GADF and Multi-Scale CNN[J]. Spectroscopy and Spectral Analysis, 2021, 41(12): 3701

    Download Citation

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

    Category: Research Articles

    Received: Dec. 26, 2020

    Accepted: --

    Published Online: Dec. 17, 2021

    The Author Email: Guo-wei YU (ygw986119533@163.com)

    DOI:10.3964/j.issn.1000-0593(2021)12-3701-07

    Topics