Beyond Pretty Pictures: Combined Single- and Multi-Image Super-resolution for Sentinel-2 Images

TL;DR

This paper introduces SEN4X, a hybrid super-resolution method that combines single- and multi-image techniques with learned priors to enhance Sentinel-2 satellite images to 2.5 m resolution, improving urban land-cover classification.

Contribution

The paper presents SEN4X, a novel hybrid super-resolution architecture that integrates temporal oversampling and learned priors from high-resolution data for Sentinel-2 images.

Findings

Significant improvement in urban land-cover classification accuracy.

Outperforms state-of-the-art super-resolution baselines.

Enhances Sentinel-2 resolution to 2.5 meters.

Abstract

Super-resolution aims to increase the resolution of satellite images by reconstructing high-frequency details, which go beyond na\"ive upsampling. This has particular relevance for Earth observation missions like Sentinel-2, which offer frequent, regular coverage at no cost; but at coarse resolution. Its pixel footprint is too large to capture small features like houses, streets, or hedge rows. To address this, we present SEN4X, a hybrid super-resolution architecture that combines the advantages of single-image and multi-image techniques. It combines temporal oversampling from repeated Sentinel-2 acquisitions with a learned prior from high-resolution Pl\'eiades Neo data. In doing so, SEN4X upgrades Sentinel-2 imagery to 2.5 m ground sampling distance. We test the super-resolved images on urban land-cover classification in Hanoi, Vietnam. We find that they lead to a significant performance improvement over state-of-the-art super-resolution baselines.