Table 9 Segmentation results on 8 real rosebush models from ROSE-X data set with ShellNet
From: Segmentation of structural parts of rosebush plants with 3D point-based deep learning methods
ShellNet | I | II | III | S | S + I | S + II | S + III | |
|---|---|---|---|---|---|---|---|---|
Re | Flower | 48.62 | 44.32 | 51.54 | 97.27 | 68.13 | 51.12 | 58.95 |
Leaf | 96.52 | 97.39 | 98.09 | 66.19 | 95.74 | 98.62 | 98.62 | |
Stem | 30.83 | 54.11 | 59.87 | 41.85 | 58.17 | 66.79 | 73.48 | |
Pr | Flower | 64.54 | 82.61 | 85.05 | 20.06 | 74.14 | 87.63 | 91.26 |
Leaf | 87.85 | 90.35 | 91.46 | 94.40 | 92.59 | 92.24 | 93.38 | |
Stem | 66.20 | 71.80 | 80.24 | 70.06 | 71.54 | 85.15 | 87.90 | |
IoU | Flower | 38.37 | 40.54 | 47.26 | 19.95 | 55.05 | 47.67 | 55.80 |
Leaf | 85.16 | 88.22 | 89.86 | 63.69 | 88.93 | 91.07 | 92.18 | |
Stem | 26.64 | 44.63 | 52.18 | 35.50 | 47.24 | 59.82 | 66.73 | |
MIoU | 50.05 | 57.80 | 63.10 | 39.71 | 63.74 | 69.19 | 71.57 | |
Acc | 85.37 | 88.43 | 90.20 | 65.72 | 89.35 | 91.40 | 92.75 | |