Super-Resolving Commercial Satellite Imagery Using Realistic Training Data
This work addresses the specific issue of enhancing commercial satellite imagery quality for applications like remote sensing, but it is incremental as it builds on existing super-resolution methods.
The paper tackled the problem of poor super-resolution performance on real satellite images by proposing a realistic training data generation model that includes both satellite imaging and ground post-processing, resulting in improved super-resolution performance on real satellite images.
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.