Table 3 Table presenting the numerical comparison on the CHN6-CUG dataset, with units in percentage (%). The top-performing values are marked in bold, while those ranking second are underlined.
From: Heterogeneous dual-decoder network for road extraction in remote sensing images
Method | Source | IoU | Precision | Recall | F1 | mIoU |
|---|---|---|---|---|---|---|
D-LinkNet15 | CVPR 18 | 60.98 | 78.13 | 71.67 | 74.76 | 77.78 |
NL-LinkNet16 | GRSL 21 | 63.52 | 76.98 | 78.41 | 77.69 | 79.12 |
CoANet36 | TIP 21 | 65.68 | 79.84 | 78.74 | 79.29 | 80.43 |
DBRANet24 | GRSL 22 | 62.07 | 78.87 | 72.53 | 75.56 | 78.42 |
DDU-Net18 | TGRS 22 | 65.18 | 79.50 | 78.36 | 78.92 | 80.13 |
DSCNet20 | ICCV 23 | 64.12 | 78.60 | 75.60 | 77.07 | 79.55 |
CFRNet30 | GRSL 24 | 64.19 | 79.72 | 76.70 | 78.18 | 79.58 |
OARENet39 | TGRS 24 | 66.18 | 80.17 | 79.13 | 79.65 | 80.75 |
Swin-GAT42 | RS 25 | 65.99 | 80.70 | 78.35 | 79.51 | 80.62 |
CGCNet43 | TGRS 25 | 66.04 | 80.35 | 76.58 | 78.42 | 80.68 |
HDDNet | Ours | 67.27 | 81.37 | 79.52 | 80.44 | 81.36 |
