Generating digital surface models
The DSM-OPT service generates digital surface models (DSMs) from Pleiades stereo and tri-stereo imagery.
DSM-OPT was developed by the EOST school and observatory of Earth sciences for the ForM@Ter data and services hub in collaboration with Theia and DINAMIS, with contributions from the MATIS stereo image analysis and processing methods research laboratory at France’s national mapping, survey and forestry agency l’IGNand the IPGPEarth physics institute, Paris. It is deployed on the A2S processing infrastructure hosted at the University of Strasbourg Mesocentre. DSM-OPT was set up with support from ESA, CNES and CNRS/INSU.
DSM-OPT can be activated from DINAMIS when searching and selecting Pleaides stereopairs or tri-stereopairs. A tutorial shows you how to search and select stereopairs or tri-stereopairs, and then how to activate the DSM processing service.
DSM-OPT can also be used from ForM@Ter after uploading imagery from your computer.
The service offers the ability for non-expert communities to generate DSMs with pre-set parameters while allowing expert communities to define more advanced parameters themselves. It is underpinned by the MicMac photogrammetry library (IGN/MATIS) and ad-hoc developments by EOST and IPGP.
In 2021, CNES opened access to the entire archive of imagery from SPOT-1 to SPOT-5 (1986-2015) in raw format (Level 1A) under its SPOT World Heritage (SWH) programme. This archive imagery can be downloaded free of charge via an online catalogue (you must create a user account if you don’t already have one and agree to the terms of the Etalab 2.0 licence).
During the course of 2022, all SWH imagery at CNES will be directly downloadable from the DINAMIS Catalogue.
In addition to this coverage, CNES provides a tool for geometric processing of SWH images using orthorectified images from ESA’s Sentinel-2 satellites to convert them from raw Level 1A format to Level 2A map format (SWH-2A-Carto).
This tool lets you upload SWH images, correct their geometry and then resample them in a map projection at constant elevation with the simplified geometric model of the original raw data, without geocoding or a digital elevation model (DEM), and without radiometric or atmospheric corrections.
SWH images converted to Level 2A are 8-bit encoded in each spectral band, so they are compatible with commonly used geographic reference systems. The tool is open and free to use.
Ultimately, the tool’s designers envision developing an online processing service able to directly convert Level 1A imagery to Level 2A before downloading.
DINAMIS will help to scale up this service and will host it in its catalogue.