Table 2 Details of the land-cover types and the number of training, validation, and testing samples for UP and HU data sets.
From: Adaptive pixel attention network for hyperspectral image classification
No. | University of Pavia (UP) (1%,0.1%,98.99%) | University of Houston (HU) (1%,0.1%,98.99%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Class | Col | Train | Val | Test | Class | Col | Train | Val | Test | |
1 | Asphalt |
| 66 | 6 | 6559 | Grass-heal |
| 12 | 1 | 1238 |
2 | Meadows |
| 186 | 18 | 18445 | Grass-stre |
| 12 | 1 | 1241 |
3 | Gravel |
| 20 | 2 | 2077 | Grass-synth |
| 6 | 1 | 690 |
4 | Trees |
| 30 | 3 | 3031 | Tree |
| 12 | 1 | 1231 |
5 | Metal |
| 13 | 1 | 1331 | Soil |
| 12 | 1 | 1229 |
6 | Bare-soil |
| 50 | 5 | 4974 | Water |
| 3 | 1 | 321 |
7 | Bitumen |
| 13 | 1 | 1316 | Residential |
| 12 | 1 | 1255 |
8 | Bricks |
| 36 | 3 | 3643 | Commercial |
| 12 | 1 | 1231 |
9 | Shadows |
| 9 | 1 | 937 | Road |
| 12 | 1 | 1239 |
10 | Â | Â | Â | Â | Â | Highway |
| 12 | 1 | 1214 |
11 | Â | Â | Â | Â | Â | Railway |
| 12 | 1 | 1222 |
12 | Â | Â | Â | Â | Â | Parking-lot1 |
| 12 | 1 | 1220 |
13 | Â | Â | Â | Â | Â | Parking-lot2 |
| 4 | 1 | 464 |
14 | Â | Â | Â | Â | Â | Tennis-court |
| 4 | 1 | 423 |
15 | Â | Â | Â | Â | Â | Running-tr |
| 6 | 1 | 653 |
| Â | Total | Â | 423 | 40 | 42313 | Total | Â | 143 | 13 | 14871 |
