earthflow/enmap_corine

How to use it

Install Dataset4EO

git clone --branch streaming https://github.com/EarthNets/Dataset4EO.git

pip install -e .

Then download the dataset from this Huggingface repo.

import dataset4eo as eodata
import time

train_dataset = eodata.StreamingDataset(input_dir="optimized_enmap_corine", num_channels=202, channels_to_select=[0,1,2], shuffle=True, drop_last=True)
sample = dataset[101]
print(sample.keys())
print(sample["image"])        
print(sample["simage"].shape)
print(sample["label"]) 

The land cover types of the CORINE dataset:

#	Code	Land Cover Type
0	1.1.1	Continuous urban fabric
1	1.1.2	Discontinuous urban fabric
2	1.2.1	Industrial or commercial units
3	1.2.2	Road and rail networks and associated land
4	1.2.3	Port areas
5	1.2.4	Airports
6	1.3.1	Mineral extraction sites
7	1.3.2	Dump sites
8	1.3.3	Construction sites
9	1.4.1	Green urban areas
10	1.4.2	Sport and leisure facilities
11	2.1.1	Non-irrigated arable land
12	2.1.2	Permanently irrigated land
13	2.1.3	Rice fields
14	2.2.1	Vineyards
15	2.2.2	Fruit trees and berry plantations
16	2.2.3	Olive groves
17	2.3.1	Pastures
18	2.4.1	Annual crops associated with permanent crops
19	2.4.2	Complex cultivation patterns
20	2.4.3	Land principally occupied by agriculture, with significant areas of natural vegetation
21	2.4.4	Agro-forestry areas
22	3.1.1	Broad-leaved forest
23	3.1.2	Coniferous forest
24	3.1.3	Mixed forest
25	3.2.1	Natural grassland
26	3.2.2	Moors and heathland
27	3.2.3	Sclerophyllous vegetation
28	3.2.4	Transitional woodland/shrub
29	3.3.1	Beaches, dunes, sands
30	3.3.2	Bare rock
31	3.3.3	Sparsely vegetated areas
32	3.3.4	Burnt areas
33	3.3.5	Glaciers and perpetual snow
34	4.1.1	Inland marshes
35	4.1.2	Peatbogs
36	4.2.1	Salt marshes
37	4.2.2	Salines
38	4.2.3	Intertidal flats
39	5.1.1	Water courses
40	5.1.2	Water bodies
41	5.2.1	Coastal lagoons
42	5.2.2	Estuaries
43	5.2.3	Sea and ocean

We acknowledge and give full credit to the original authors of SpectralEarth for their effort in creating this dataset. The dataset is re-hosted in compliance with its original license to facilitate further research. Please cite the following paper for the creation of the dataset:

@article{braham2024spectralearth,
  title={SpectralEarth: Training Hyperspectral Foundation Models at Scale},
  author={Braham, Nassim Ait Ali and Albrecht, Conrad M and Mairal, Julien and Chanussot, Jocelyn and Wang, Yi and Zhu, Xiao Xiang},
  journal={arXiv preprint arXiv:2408.08447},
  year={2024}
}