DIONE - Super Resolution using Sentinel-2 and Deimos imagery

This repo contains code to train a multitemporal super-resolution model for Sentinel-2 imagery using Deimos.

You can find more information about this project in the blog post Multi-temporal Super-Resolution on Sentinel-2 Imagery

Introduction

This project is part of the DIONE project where one of the missions is using novel techniques to improve the capabilities of satellite technology while integrating various data sources, such as very high resolution imagery, to, for example, enable monitoring of smaller agricultural parcels through the use of super resolution models.

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 870378.

Please visit the Dione website for further information.

Requirements

The super resolution pipeline uses SentinelHub service to download Sentinel-2 and Deimos imagery. Amazon AWS S3 bucket was used to store the data.
Deimos imagery is not public, however any other Very High Resolution imagery can be used by adjusting the general workflow.

Installation and usage

To install the sr package, clone locally the repository, and from within the repository, run the following commands:

pip install -r requirements.txt
python setup.py install --user

Procedure is executed in notebooks, the basic functionality of each notebook is described below:

• 00-parse-deimos-metadata.ipynb: Deimos specific. Parses metadata for each ingested Deimos tile and saves to dataframe.
• 00a-add-per-tile-median.ipynb Deimos specific. Calculates median for each Deimos tile.
• 00b-calculate-cloudfree-deimos-stats.ipynb Deimos specific. Calculates Deimos tile statistics on cloudless areas.
• 01-download-to-eopatches.ipynb Download Sentinel-2 and ingested Deimos imagery to EOPatches
• 02a-add-clm-deimos.ipynb Add cloud mask information to Deimos EOPatches
• 02b-add-clm-stats-to-patches.ipynb Add cloudless normalization statistics to EOPatches
• 03-sampling.ipynb Sample smaller patchlets from EOPatches
• 04-sampled-to-npz.ipynb Construct NPZ files from patchlets.
• 05a-train-test-split.ipynb Split NPZ files into train/test/validation sets..
• 05b-find-cloudy-neighbours.ipynb Shadow detection by filtering neighbours of cloudy EOPatches
• 05c-calculate-s2-normalizations.ipynb Calculate per country Sentinel-2 normalization statistics.
• 06-train.ipynb Model training.
• 07-predict.ipynb Predict the model on smaller patchlets.
• 07b-predict-eopatches.ipynb Predict the model on whole EOPatches.

Acknowledgments

The code is adapted from ElementAI's HighResNet code. Refer also to the published paper.