• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 11, 3568 (2022)
Shi-jiao GAO1、*, Hai-ou GUAN1、1; *;, Xiao-dan MA1、1;, and Yan-hong WANG2、2;
Author Affiliations
  • 11. College of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, China
  • 22. College of Horticulture and Landscape Architecture, Heilongjiang Bayi Agricultural University, Daqing 163319, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2022)11-3568-07 Cite this Article
    Shi-jiao GAO, Hai-ou GUAN, Xiao-dan MA, Yan-hong WANG. Soybean Canopy Extraction Method Based on Multispectral Image Processing[J]. Spectroscopy and Spectral Analysis, 2022, 42(11): 3568 Copy Citation Text show less
    Original images of soybean(a): GRE; (b): NIR; (c): RED; (d): REG; (e): RGB
    Fig. 1. Original images of soybean
    (a): GRE; (b): NIR; (c): RED; (d): REG; (e): RGB
    Comparison of filtering effects of multi-spectral images(a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Fig. 2. Comparison of filtering effects of multi-spectral images
    (a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Multispectral images of soybean canopy with iteration method(a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Fig. 3. Multispectral images of soybean canopy with iteration method
    (a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Multispectral images of soybean canopy by Otsu method(a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Fig. 4. Multispectral images of soybean canopy by Otsu method
    (a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Multispectral images of soybean canopy by local threshold method(a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Fig. 5. Multispectral images of soybean canopy by local threshold method
    (a1): GRE; (a2): NIR; (a3): RED; (a4): REG; (a5): RGB;(b1): GRE; (b2): NIR; (b3): RED; (b4): REG; (b5): RGB
    Entropy statistics of segmented images
    Fig. 6. Entropy statistics of segmented images
    函数指标
    PSNRSSIM
    GRENIRREDREGRGBGRENIRREDREGRGB
    均值滤波34.71833.35430.03140.62137.0680.9350.9370.8480.9610.970
    高斯滤波36.18134.85731.40241.99138.6380.9510.9510.8890.9700.978
    Table 1. Multispectral image preprocessing evaluation index
    图像类型分割算法
    迭代法Otsu法局部阈值法
    有效分割过分割欠分割有效分割过分割欠分割有效分割过分割欠分割
    GRE0.568 60.533 80.284 70.871 40.222 90.191 20.870 60.227 70.206 7
    NIR0.978 10.113 70.136 10.944 00.116 80.135 30.987 50.079 10.202 4
    RED0.052 40.967 10.288 20.130 10.913 20.142 30.167 50.885 80.123 6
    REG0.691 60.366 80.229 20.975 50.094 40.227 00.981 50.090 50.227 3
    RGB0.879 90.202 20.632 50.828 10.270 40.456 40.859 80.228 70.505 0
    Table 2. Comparison of evaluation indexes of different segmentation algorithms
    Shi-jiao GAO, Hai-ou GUAN, Xiao-dan MA, Yan-hong WANG. Soybean Canopy Extraction Method Based on Multispectral Image Processing[J]. Spectroscopy and Spectral Analysis, 2022, 42(11): 3568
    Download Citation