Remote Sensing Single Image Super-Resolution Benchmarking with Transfer Learning Algorithms

In the context of real-world applications like medical imaging systems, tracking, astronomical imaging, navigation, and remote sensing (RS), there is a pressing need to enhance or upscale images with minimal errors. This is particularly critical for tasks such as target detection, image classification, and land use mapping. However, remote sensing images often suffer from limitations in spatial, spectral, radiometric, and temporal resolution due to complex atmospheric conditions and sensor constraints. Additionally, acquiring these images can be expensive and time-consuming. In this study, we propose a Single Image Super-Resolution (SISR) method to address these challenges by upscaling low-quality remote sensing images to higher resolution, enabling a better understanding of these images. We also discuss the specific challenges in remote sensing super-resolution techniques and review various upscaling approaches, while analyzing the impact of other factors like weather conditions, image capture time, and different scene types on the technique’s effectiveness.