• Spectroscopy and Spectral Analysis
  • Vol. 41, Issue 1, 216 (2021)
Xiao-dan MA1、1、*, Meng LIU1、1, Hai-ou GUAN1、1, Feng-rui WEN1、1, and Gang LIU1、1
Author Affiliations
  • 1[in Chinese]
  • 11. College of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2021)01-0216-07 Cite this Article
    Xiao-dan MA, Meng LIU, Hai-ou GUAN, Feng-rui WEN, Gang LIU. Recognition Method for Crop Canopies Based on Thermal Infrared Image Processing Technology[J]. Spectroscopy and Spectral Analysis, 2021, 41(1): 216 Copy Citation Text show less
    Double-source images(a): Thermal infrared image; (b): Visible image
    Fig. 1. Double-source images
    (a): Thermal infrared image; (b): Visible image
    Visible image segmentation(a): Visible image of sample;(b): Visible image of canopy after segmentation
    Fig. 2. Visible image segmentation
    (a): Visible image of sample;(b): Visible image of canopy after segmentation
    Feature points match images
    Fig. 3. Feature points match images
    Initial registration effect of thermal infrared image and reference image(a): Original thermal infrared image;(b): Initial registration effect
    Fig. 4. Initial registration effect of thermal infrared image and reference image
    (a): Original thermal infrared image;(b): Initial registration effect
    Image exact matching process(a): Angle registration; (b): Horizontal registration; (c): Vertical registration; (d): Proportional registration
    Fig. 5. Image exact matching process
    (a): Angle registration; (b): Horizontal registration; (c): Vertical registration; (d): Proportional registration
    Contrast of recognition effect of thermal infrared images(a): Original thermal infrared image; (b): Energy distribution of original image;(c): Recognition effect of thermal infrared image of canopy; (d): Energy distribution of canopy image after recognition
    Fig. 6. Contrast of recognition effect of thermal infrared images
    (a): Original thermal infrared image; (b): Energy distribution of original image;(c): Recognition effect of thermal infrared image of canopy; (d): Energy distribution of canopy image after recognition
    Mutual information contrast of recognition effect of thermal infrared images
    Fig. 7. Mutual information contrast of recognition effect of thermal infrared images
    可调参
    数因子
    调整范围温度/℃熵值
    最小值最大值平均值
    θ(-1~1)16.3519.7118.028 62.174 4
    tx(0~35)16.3519.7518.055 92.171 3
    ty(-45~30)16.3519.6317.986 62.179 6
    ρ(0.80~0.90)16.3519.5217.839 32.031 4
    Table 1. Changes of temperature and entropy
    序号有效分割过分割欠分割分割图像熵值
    标准本文
    10.969 60.069 80.006 83.173 03.084 1
    20.971 00.066 10.006 24.196 72.976 6
    30.989 50.061 10.003 45.218 15.069 9
    40.949 90.070 70.014 85.438 95.198 7
    50.958 90.076 30.021 55.048 04.903 2
    平均0.967 80.068 80.010 54.614 94.246 5
    Table 2. Reference image evaluation index
    Xiao-dan MA, Meng LIU, Hai-ou GUAN, Feng-rui WEN, Gang LIU. Recognition Method for Crop Canopies Based on Thermal Infrared Image Processing Technology[J]. Spectroscopy and Spectral Analysis, 2021, 41(1): 216
    Download Citation