Super-Resolving Commercial Satellite Imagery Using Realistic Training Data

TL;DR

This paper introduces a realistic training data generation model and a specialized neural network to improve super-resolution of real satellite images, addressing the gap between synthetic training data and real-world application.

Contribution

The paper presents a novel training data generation approach that models the entire satellite imaging and ground processing, enhancing super-resolution performance on real images.

Findings

Improved super-resolution results on real satellite images.

The realistic training data model outperforms traditional synthetic models.

The specialized CNN achieves higher quality reconstructions.

Abstract

In machine learning based single image super-resolution, the degradation model is embedded in training data generation. However, most existing satellite image super-resolution methods use a simple down-sampling model with a fixed kernel to create training images. These methods work fine on synthetic data, but do not perform well on real satellite images. We propose a realistic training data generation model for commercial satellite imagery products, which includes not only the imaging process on satellites but also the post-process on the ground. We also propose a convolutional neural network optimized for satellite images. Experiments show that the proposed training data generation model is able to improve super-resolution performance on real satellite images.