Paper: J. Li, X. Huang, and L. Tu, “WHU-OHS : A benchmark dataset for large-scale Hersepctral Image classification,” Int. J. Appl. Earth Obs. Geoinf., vol. 113, no. September, p. 103022, 2022, doi: 10.1016/j.jag.2022.103022. [Link]
Dataset download: http://irsip.whu.edu.cn/resources/WHU_OHS_show.php
The WHU-OHS dataset is made up of 42 OHS satellite images acquired from more than 40 different locations in China (Fig. 1). The imagery has a spatial resolution of 10 m (nadir) and a swath width of 60 km (nadir). There are 32 spectral channels ranging from the visible to near-infrared range, with an average spectral resolution of 15 nm. We cropped each image into 512 × 512 pixels with a stride of 32. There are 4822, 513, and 2460 sub-images in the training, validation, and test sets, respectively.
Fig. 1. Left: The geographical locations of the 42 images in the WHU-OHS dataset. Right: Examples of local OHS parcels (true-color compositions with R: 670 nm; G: 566 nm; B: 480 nm) and their corresponding reference labels.
The dataset was organized in the format shown in Fig. 2.
Fig. 2. Data organization of the WHU-OHS dataset.
The correspondence of label IDs and categories:
| ID | Category | ID | Category | ID | Category | ID | Category |
|---|---|---|---|---|---|---|---|
| 1 | Paddy field | 7 | High-covered grassland | 13 | Beach land | 19 | Gobi |
| 2 | Dry farm | 8 | Medium-covered grassland | 14 | Shoal | 20 | Saline-alkali soil |
| 3 | Woodland | 9 | Low-covered grassland | 15 | Urban built-up | 21 | Marshland |
| 4 | Shrubbery | 10 | River canal | 16 | Rural settlement | 22 | Bare land |
| 5 | Sparse woodland | 11 | Lake | 17 | Other construction land | 23 | Bare rock |
| 6 | Other forest land | 12 | Reservoir pond | 18 | Sand | 24 | Ocean |
Pytorch toolbox for large-scale hyperspectral image classification using WHU-OHS dataset. The deep network models will be updated continuously.
Update:
2024.3.11 Example code for transfer learning from WHU-OHS dataset to public hyperspectral datasets is updated. See the "transfer" folder.
Deep network models:
1D-CNN: W. Hu, Y. Huang, L. Wei, F. Zhang, and H. Li, “Deep Convolutional Neural Networks for Hyperspectral Image Classification,” J. Sensors, vol. 2015, p. 12, 2015.
3D-CNN: Y. Chen, H. Jiang, C. Li, X. Jia, and P. Ghamisi, “Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural Networks,” IEEE Trans. Geosci. Remote Sens., vol. 54, pp. 6232–6251, 2016.
A2S2K-ResNet: S. K. Roy, S. Manna, T. Song, and L. Bruzzone, “Attention-Based Adaptive Spectral–Spatial Kernel ResNet for Hyperspectral Image Classification,” IEEE Trans. Geosci. Remote Sens., vol. 59, no. 9, pp. 7831–7843, 2021.
FreeNet: Z. Zheng, Y. Zhong, A. Ma, and L. Zhang, “FPGA: Fast Patch-Free Global Learning Framework for Fully End-to-End Hyperspectral Image Classification,” IEEE Trans. Geosci. Remote Sens., vol. 58, no. 8, pp. 5612–5626, 2020.
Accuracy on test set for reference (taking S1: Changchun as an example):
| Network | 1D-CNN | 3D-CNN | A2S2K-ResNet | FreeNet |
|---|---|---|---|---|
| OA | 0.636 | 0.766 | 0.809 | 0.847 |
| Kappa | 0.526 | 0.700 | 0.757 | 0.806 |
| mIoU | 0.227 | 0.305 | 0.419 | 0.480 |