Cross-sensor super-resolution of irregularly sampled Sentinel-2 time series

TL;DR

This paper explores multi-image super-resolution for irregularly sampled Sentinel-2 satellite data, demonstrating improved reconstruction quality using deep learning models and introducing a new dataset for 4x super-resolution.

Contribution

It extends existing super-resolution algorithms to handle irregular temporal sampling and introduces BreizhSR, a novel dataset for satellite image super-resolution.

Findings

Multiple images enhance super-resolution quality

Temporal positional encoding enables time-specific super-resolution

Trade-off observed between spectral fidelity and perceptual quality

Abstract

Satellite imaging generally presents a trade-off between the frequency of acquisitions and the spatial resolution of the images. Super-resolution is often advanced as a way to get the best of both worlds. In this work, we investigate multi-image super-resolution of satellite image time series, i.e. how multiple images of the same area acquired at different dates can help reconstruct a higher resolution observation. In particular, we extend state-of-the-art deep single and multi-image super-resolution algorithms, such as SRDiff and HighRes-net, to deal with irregularly sampled Sentinel-2 time series. We introduce BreizhSR, a new dataset for 4x super-resolution of Sentinel-2 time series using very high-resolution SPOT-6 imagery of Brittany, a French region. We show that using multiple images significantly improves super-resolution performance, and that a well-designed temporal positional encoding allows us to perform super-resolution for different times of the series. In addition, we observe a trade-off between spectral fidelity and perceptual quality of the reconstructed HR images, questioning future directions for super-resolution of Earth Observation data.