Enhancing PROBA-V Satellite Imagery for Vegetation Monitoring Using FSRCNN-Based Super-Resolution

Super-resolution techniques play a crucial role in improving the quality of photographs, especially those taken with lesser resolution or from a distance using cameras or sensors. Improving these images brings a huge improvement in their clarity and fine details, which directly affects the usefulness and worth of remote sensing data products. Based on the available input data, image super-resolution techniques can be divided into two main types. The goal of single-image super-resolution is to convert one low-resolution picture to its equivalent high-resolution image. Multi-image super-resolution, on the other hand, uses a set of low-resolution photos that capture the same object or scene in order to produce a higher-resolution image by mapping the original low-resolution photos. Both techniques make use of generative adversarial networks (GANs) to create enhanced or completely new images and convolutional neural networks (CNNs) to extract important image attributes, playing pivotal roles in this enhancement process.