Super-resolution Pléiades/SPOT

Super-resolution Pléiades/SPOT quadruples the image resolution of Pléiades and SPOT satellite imagery to increase the performance of object detection algorithms.

Supported workflows

Data platform:

• Catalog -> Pléiades Analytic or SPOT 6/7 Analytic -> Order and get the asset from storage of UP42
• In Projects workflows, Processing from Storage -> Pan-sharpening -> Super-resolution Pléiades/SPOT

Remark: Pan-sharpening includes the necessary DIMAP to GeoTIFF conversion.

Algorithm

Super-resolution is the process of increasing the resolution of images using an algorithm. This block is based on a state-of-the-art Convolutional Neural Network algorithm. Quality improvements of the images were measured using the SSIM metric to guarantee that information content was added by the algorithm.

The image resolution of the processed images will be quadrupled, but the information content of algorithmically derived images does not contain more information than images that were originally recorded at a higher resolution.

Use Case

The use case for this block is as a preprocessing step for object detection algorithms (ships, cars, planes etc) as the images become much crisper and contour outlines more well-defined.

Note: The runtime of the algorithm is between 30 and 60 minutes per sqkm for Pléiades imagery.

More information

Click here to learn more about the block’s documentation.