Spectroscopy and Spectral Analysis, Volume. 40, Issue 12, 3865(2020)

Visible/Near-Infrared Spectroscopy Combined With Machine Vision for Dynamic Detection of Aflatoxin B1 Contamination in Peanut

Chen YAN1,1, Xue-song JIANG1,1, Fei SHEN1,1, Xue-ming HE1,1, Yong FANG1,1, Qin LIU1,1, Hong-ping ZHOU1,1, and Xing-quan LIU1,1
Author Affiliations
  • 1[in Chinese]
  • 11. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • show less
    Figures & Tables(8)
    Visible/near infrared spectroscopy and machine vision inspection system1: Drive motor; 2: Keyboard; 3: Slide plate; 4: Diffuse reflection probe; 5: displayer; 6: Optical fiber; 7: Spectrograph; 8: Camera obscura; 9: Digital camera; 10: Ring light source; 11: Conveyor
    Average raw (a) and second derivative (b) spectra of peanut samples contaminated five different levels of AFB1
    Original and gray scale images of peanut samples contaminated different levels of AFB1
    RGB value of peanut samples contaminated with different levels of AFB1
    Weighting coefficients for PC1—PC3 of peanut samples
    Score plot of PC1 and PC2 of peanut samples(a): Visible/near-infrared full-band; (b): Characteristic wavelength;(c): Image; (d): Spectral and image fusion characteristic parameters
    • Table 1. Results of LDA classification model of peanut samples based on different data forms

      View table
      View in Article

      Table 1. Results of LDA classification model of peanut samples based on different data forms

      模型名称数据形式准确率/%
      OriginalSNVMSCBaselineS-G
      建模集预测集建模集预测集建模集预测集建模集预测集建模集预测集
      LDA全谱段
      特征波长
      图像
      融合
      94
      88
      86
      96
      90
      88
      84
      86
      94
      89

      92
      90
      83

      86
      94
      89

      95
      92
      86

      86
      94
      85

      94
      90
      74

      86
      93
      88

      96
      92
      82

      86
    • Table 2. Results of SVM classification model of peanut samples based on different data forms

      View table
      View in Article

      Table 2. Results of SVM classification model of peanut samples based on different data forms

      模型名称数据形式准确率/%
      PolynomialRadial basis functionSigmoid
      建模集预测集建模集预测集建模集预测集
      Nu-SVM全谱段
      特征波长
      图像
      融合
      66
      65
      89
      97
      54
      54
      86
      90
      89
      89
      96
      94
      82
      80
      90
      92
      85
      89
      75
      87
      74
      78
      64
      74
      C-SVM全谱段
      特征波长
      图像
      融合
      65
      76
      83
      84
      50
      50
      78
      78
      87
      86
      94
      95
      76
      74
      88
      92
      85
      82
      76
      85
      74
      74
      50
      72
    Tools

    Get Citation

    Copy Citation Text

    Chen YAN, Xue-song JIANG, Fei SHEN, Xue-ming HE, Yong FANG, Qin LIU, Hong-ping ZHOU, Xing-quan LIU. Visible/Near-Infrared Spectroscopy Combined With Machine Vision for Dynamic Detection of Aflatoxin B1 Contamination in Peanut[J]. Spectroscopy and Spectral Analysis, 2020, 40(12): 3865

    Download Citation

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

    Category: Research Articles

    Received: Oct. 27, 2019

    Accepted: --

    Published Online: Jun. 18, 2021

    The Author Email:

    DOI:10.3964/j.issn.1000-0593(2020)12-3865-06

    Topics