• Laser & Optoelectronics Progress
  • Vol. 62, Issue 2, 0228002 (2025)
Yulin Cai1,*, Hongzhen Gao1, Xiaole Fan1, Huiyu Xu1..., Zhengjun Liu2 and Geng Zhang2|Show fewer author(s)
Author Affiliations
  • 1College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, Shandong , China
  • 2Chinese Academy of Surveying and Mapping, Beijing 100036, China
  • show less
    DOI: 10.3788/LOP241175 Cite this Article Set citation alerts
    Yulin Cai, Hongzhen Gao, Xiaole Fan, Huiyu Xu, Zhengjun Liu, Geng Zhang. Fine Classification of Tree Species Based on Improved U-Net Network[J]. Laser & Optoelectronics Progress, 2025, 62(2): 0228002 Copy Citation Text show less
    Channel attention mechanism
    Fig. 1. Channel attention mechanism
    CA-U-Net structure
    Fig. 2. CA-U-Net structure
    Location of study areas
    Fig. 3. Location of study areas
    LiDAR point cloud data and true-color images for study area 1 and 2. (a) LiDAR point cloud data for study area 1; (b) LiDAR point cloud data for study area 2; (c) true-color image for study area 1; (d) true-color image for study area 2
    Fig. 4. LiDAR point cloud data and true-color images for study area 1 and 2. (a) LiDAR point cloud data for study area 1; (b) LiDAR point cloud data for study area 2; (c) true-color image for study area 1; (d) true-color image for study area 2
    Comparison of classification results of CA-U-Net and U-Net
    Fig. 5. Comparison of classification results of CA-U-Net and U-Net
    Model training process. (a) U-Net; (b) CA-U-Net
    Fig. 6. Model training process. (a) U-Net; (b) CA-U-Net
    Proportion of different samples
    Fig. 7. Proportion of different samples
    Comparison of classification results of CB-CA-U-Net and CA-U-Net
    Fig. 8. Comparison of classification results of CB-CA-U-Net and CA-U-Net
    ClassificationFCNSegNetU-NetCA-U-Net
    R /%F1 /%R /%F1 /%R /%F1 /%R /%F1 /%
    Silver birch97.2682.5195.9586.3598.0297.4395.1096.52
    Camphor49.6663.6762.6857.0399.9298.8799.9299.20
    Beech97.0194.2197.5096.3996.6296.1698.0496.88
    Hackberry64.2575.8913.7122.1188.7692.9491.4493.82
    Banyan tree93.3294.6096.4691.3998.3591.2498.0995.66
    Mahogany96.6595.4192.0495.3979.0387.5695.4397.32
    Small-leaved olive97.8688.2997.5696.4898.1897.9298.2098.95
    OA /%88.2484.8193.4996.80
    Kappa0.860.820.920.96
    Table 1. Comparison of classification accuracy between different networks
    ClassificationCA-U-NetCB-CA-U-Net
    R /%F1 /%R /%F1 /%
    Camellia oleifera91.6277.7581.3782.39
    Chinese fir89.3387.2485.2290.57
    Small-leaved olive67.0777.2468.2073.04
    Pinus massoniana92.0786.9190.5991.25
    Bamboo54.1357.3771.7667.94
    Camphor83.7685.1191.2384.05
    Beech65.1565.5280.8782.39
    OA /%83.9387.96
    Kappa0.800.85
    Table 2. Comparison of performance of CA-U-Net and CB-CA-U-Net
    Yulin Cai, Hongzhen Gao, Xiaole Fan, Huiyu Xu, Zhengjun Liu, Geng Zhang. Fine Classification of Tree Species Based on Improved U-Net Network[J]. Laser & Optoelectronics Progress, 2025, 62(2): 0228002
    Download Citation