PROBA-V Super-Resolution dataset

<p>The PROBA-V Super-Resolution dataset is the official dataset of&nbsp;<strong>ESA&#39;s Kelvins</strong>&nbsp;<strong>competition for &quot;PROBA-V Super Resolution&quot;</strong>. It contains satellite data from 74 hand-selected regions around the globe at different points in time.&nbsp;The data is composed of radiometrically and geometrically corrected Top-Of-Atmosphere (TOA) reflectances for the RED and NIR spectral bands at&nbsp;<strong>300m</strong>&nbsp;and&nbsp;<strong>100m</strong>&nbsp;resolution in Plate Carr&eacute;e projection. The&nbsp;<strong>300m</strong>&nbsp;resolution data is delivered as&nbsp;<strong>128x128</strong>&nbsp;grey-scale pixel images, the&nbsp;<strong>100m</strong>&nbsp;resolution data as&nbsp;<strong>384x384</strong>&nbsp;grey-scale pixel images. Additionally, a quality map is provided for each pixel, indicating whether the pixels are concealed (i.e. by clouads, ice, water, missing information, etc.).</p> <p>The goal of the challenge can be described as <strong>Multi-Image Super-resolution</strong>: Construct a single high-resolution image out of a series of more frequent low resolution images.</p> <p>Detailed information about the related competition can be found at&nbsp;<a href="https://kelvins.esa.int/proba-v-super-resolution">https://kelvins.esa.int/proba-v-super-resolution</a>.</p> <p>A publication about the generation of this dataset exists as well:&nbsp;</p> <ul> <li><strong>M&auml;rtens&nbsp;M., Izzo D., Krzic A. and Cox D.</strong>&nbsp;&quot;Super-resolution of PROBA-V images using convolutional neural networks.&quot; Astrodynamics 3.4 (2019): 387-402. (<a href="https://arxiv.org/pdf/1907.01821.pdf">arxiv version</a>)</li> </ul>