An Open Hyperspectral Dataset with Sea-Land-Cloud Ground-Truth from the HYPSO-1 Satellite

TL;DR

This paper introduces a comprehensive open hyperspectral dataset from the HYPSO-1 satellite, including pixel-level labels for sea, land, and cloud categories, and demonstrates its utility with a deep learning model outperforming existing methods.

Contribution

The creation of a large, labeled hyperspectral dataset with ground-truth for sea, land, and cloud, and the demonstration of its effectiveness with a new deep learning approach.

Findings

Deep learning model achieved state-of-the-art performance.

Dataset includes 200 hyperspectral images with 25 million labeled spectral signatures.

Open access to dataset, labels, and code for scientific research.

Abstract

Hyperspectral Imaging, employed in satellites for space remote sensing, like HYPSO-1, faces constraints due to few labeled data sets, affecting the training of AI models demanding these ground-truth annotations. In this work, we introduce The HYPSO-1 Sea-Land-Cloud-Labeled Dataset, an open dataset with 200 diverse hyperspectral images from the HYPSO-1 mission, available in both raw and calibrated forms for scientific research in Earth observation. Moreover, 38 of these images from different countries include ground-truth labels at pixel-level totaling about 25 million spectral signatures labeled for sea/land/cloud categories. To demonstrate the potential of the dataset and its labeled subset, we have additionally optimized a deep learning model (1D Fully Convolutional Network), achieving superior performance to the current state of the art. The complete dataset, ground-truth labels, deep learning model, and software code are openly accessible for download at the website https://ntnu-smallsat-lab.github.io/hypso1_sea_land_clouds_dataset/ .