Selected Presentations
Home > Presentations > Abstract details
Title
OPERATIONAL USE OF THE ORFEO TOOL BOX FOR THE VENUS MISSION
Abstract
The Orfeo Tool Box open source software has been selected by the French Space Agency (CNES) for the development of the Venus ground segment L2 and L3 processors.
The Venus (Vegetation & Environment new micro satellite) mission is a French-Israeli joint project designed to operate a multispectral instrument for imaging selected scientific sites with a high revisit frequency (2 days). The scientific mission aims at providing the principal investigators with high spatial and spectral resolution data for environment monitoring (5m/ 12 bands). In addition the Venus instrument offers stereoscopic capabilities for cloud detection and DEM generation.
OTB provides the image processing framework on which the L2 and L3 processors are being developed. The L2 processor offers advanced atmospheric correction algorithms including water vapour and aerosol estimates based on multi-temporal data analysis. It also provides cloud mask generation and biophysical mapping.
The L3 processor is used to merge L2 time series over short periods (~10 days) in order to remove cloudy regions and generate useful syntheses for downstream users.
OTB provides built-in and highly optimized streaming and tiling capabilities from ITK (Insight ToolKit). This is a considerable asset given the data volume to be processed : a typical L1 ortho-rectified image for Venus is 8000 X 8000 pixels X 12 bands. OTB library also provides a framework for atmospheric correction and data fusion which is used to integrate Venus-specific algorithms. These methods may be extended for future multi-spectral missions such as Sentinel-2.
So far OTB has been extensively used in research projects. The Venus L2 and L3 processors represent the first use of OTB in an operational environment. This project gives us a great opportunity to demonstrate OTB performance as well as its usefulness for the development of instrument processing facilities.
Authors
Thomas Feuvrier - CS-SI
Cyrille Valladeau - CS-SI
