• Spectroscopy and Spectral Analysis
  • Vol. 41, Issue 8, 2598 (2021)
Ai-guo OUYANG*, Hao-chen LIU, Long CHENG, Xiao-gang JIANG, Xiong LI, and Xuan HU
Author Affiliations
  • School of Mechatronics & Vehicle Engineering, East China JiaoTong University, National and Local Joint Engineering Research Center of Fruit Intelligent Photoelectric Detection Technology and Equipment, Nanchang 330013, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2021)08-2598-06 Cite this Article
    Ai-guo OUYANG, Hao-chen LIU, Long CHENG, Xiao-gang JIANG, Xiong LI, Xuan HU. Hyperspectral Image Features Combined With Spectral Features Used to Classify the Bruising Time of Peach[J]. Spectroscopy and Spectral Analysis, 2021, 41(8): 2598 Copy Citation Text show less
    Hyperspectral image of experimental samples
    Fig. 1. Hyperspectral image of experimental samples
    Sketch map of hyperspectral system
    Fig. 2. Sketch map of hyperspectral system
    Spectra of samples without bruise and with different bruise time
    Fig. 3. Spectra of samples without bruise and with different bruise time
    The first five PC images obtained by PCA
    Fig. 4. The first five PC images obtained by PCA
    Weight coefficient of PC1 image
    Fig. 5. Weight coefficient of PC1 image
    Feature images of samples with different bruising time
    Fig. 6. Feature images of samples with different bruising time
    Gray histogram statistical feature extraction process
    Fig. 7. Gray histogram statistical feature extraction process
    Score projection of the first three principal components (PCs) of all peaches
    Fig. 8. Score projection of the first three principal components (PCs) of all peaches
    输入
    变量数
    碰伤时间
    /h
    RBF_KernelLin_Kernel
    误判数正确率/%误判数正确率/%
    17612680.00976.67
    17624196.67196.67
    176360100.00196.67
    176480100.000100.00
    Table 1. Results of LS-SVM modeling based on the spectral features
    输入
    变量数
    碰伤时间
    /h
    RBF_KernelLin_Kernel
    误判数正确率/%误判数正确率/%
    4121453.331936.67
    4241163.331743.33
    436583.33680.00
    448293.33583.33
    Table 2. Results of LS-SVM modeling based on the image features
    输入
    变量数
    碰伤时间
    /h
    RBF_KernelLin_Kernel
    误判数正确率/%误判数正确率/%
    18012583.331066.67
    18024196.67293.33
    180360100.00680.00
    180480100.00583.33
    Table 3. Results of LS-SVM modeling based on the image features combined with spectral features
    Ai-guo OUYANG, Hao-chen LIU, Long CHENG, Xiao-gang JIANG, Xiong LI, Xuan HU. Hyperspectral Image Features Combined With Spectral Features Used to Classify the Bruising Time of Peach[J]. Spectroscopy and Spectral Analysis, 2021, 41(8): 2598
    Download Citation