HySpecNet-11k: A large-scale hyperspectral benchmark dataset

HySpecNet-11k is a large-scale hyperspectral benchmark dataset made up of 11,483 nonoverlapping image patches acquired by the EnMAP satellite. Each patch is a portion of 128 × 128 pixels with 224 spectral bands and with a ground sample distance of 30 m.
To construct HySpecNet-11k, a total of 250 EnMAP tiles acquired during the routine operation phase between 2 November 2022 and 9 November 2022 were considered. The considered tiles are associated with less than 10% cloud and snow cover. The tiles were radiometrically, geometrically and atmospherically corrected (L2A water & land product). Then, the tiles were divided into nonoverlapping image patches. The cropped patches at the borders of the tiles were eliminated. As a result, more than 45 patches per tile are obtained, resulting in 11,483 patches for the full dataset.
We provide predefined splits obtained by randomly dividing HySpecNet into: i) a training set that includes 70% of the patches, ii) a validation set that includes 20% ..., For further details about HySpecNet-11k, please see our paper: M. H. P. Fuchs and B. Demіr, \"HySpecNet-11k: A Large-Scale Hyperspectral Dataset for Benchmarking Learning-Based Hyperspectral Image Compression Methods\", IEEE International Geoscience and Remote Sensing Symposium, Pasadena, California, 2023.,
To unpack the \"hyspecnet-11k-*.tar.gz\" tarballs use the following command: cat *.tar.gz | tar -ixzv

The \"*-DATA.npy\" files have to be generated by running the tif_to_npy.ipynb notebook from the HySpecNet Tools.
For further details visit our website at https://hyspecnet.rsim.berlin