The first part of the main processing retrieves context and search imagettes of the OLCI and SLSTR measurements respectively. This module is repeated for each OLCI camera and each GCP from the GCP database.

Extraction of the context imagette requires determination of the GCP location in OLCI geometry. Inverse ortho-geolocation of the OLCI reference channel is used to determine the context window in the OLCI image. Some checks on the window are carried out (size of the window, distance to image border, cloud flags). These tests are performed sequentially and if any of the activated tests fails, the next tests are not performed on the current imagette and the current GCP is rejected. Before extracting the radiometric information within the window, a low-pass filter is applied to the OLCI image.

The search window is defined in OLCI geometry by sliding the context window according to shift vectors. Corresponding grid points in SLSTR geometry are obtained by applying the ortho-geolocation of OLCI and subsequently the inverse ortho-geolocation of SLSTR. The SLSTR radiometric measurements at those pixels found in the SLSTR image are then interpolated on the OLCI geometry with a Shannon truncated kernel. A check for clouds in SLSTR flags is also performed at this stage.

Schematic overview of imagette extraction processing

Logical flow of extraction of OLCI/SLSTR imagette pairs module