JulienOsman
I am a research and software engineer in the fields of satellite imagery and Machine Learning. Expert in those fields after a PhD, I currently works within the CS GROUP Space BU, developing the satellite image processing library Orfeo Toolbox (OTB). I worked at the implementation of optical sensor calibration algorithms and the conception and development of image quality centers. As I was a technical support for the CNES (French Space Agency), I participated in campaigns of massive production of Sentinel2 products level L2A creation using MAJA, and in the generation of synthesis images in preparation of futures spatial missions.
Sessions
Orfeo Toolbox (OTB) is a free and open-source remote sensing software. It is available on multiple platforms, Linux, Windows and MacOs, and was developed primarily by CNES (French Space Agency) and CS Group in the frame of the development of the ORFEO program (French and Italian support program for Pleiades and Cosmo-Skymed).
OTB can process large images thanks to its built-in streaming and multithreading mechanisms. Its data processing schema is primarily based on ITK pipelines, and uses GDAL dependency to read and write raster and vector data. Many formats are supported by the library (at least those supported by GDAL) as CosmoSkyMed, Formosat, Ikonos, Pleiades, QuickBird, Radarsat 2, Sentinel 1, Spot5, Spot 6/7, TerraSarX or WorldView 2.
OTB provides a lot of applications to process optical and SAR products: ortho-rectification, calibration, pansharpening, classification, large-scale segmentation and more. The library is written in C++ but all the applications can also be accessed from Python, command line launcher, QGIS and Monterverdi, a powerful satellite image visualization tool bundled in the OTB packages capable of manipulating large images efficiently.
The library also facilitates external contributions thanks to the remote module functionality: users can add new applications without modifying the core of the library. If this new remote module is relevant, it could be added as an official remote module, like DiapOTB (differential SAR interferometric processing chain) and OTBTensorflow (multi-purpose deep learning framework, targeting remote sensing images processing).
Moreover, several operational image processing chains are based on OTB: their algorithms use the framework of OTB Applications while the orchestration is written in python. Some of the chains are also open source: Let It Snow (Snow cover detection), iota2 (Large Scale Land Surface Classification), WASP (Multitemp images fusion), S1Tiling (Sentinel-1 calibration and MAJA (Maccs-Atcor Joint Algorithm). The Orfeo Toolbox is also a part of the Sentinel 2 ground segment, being integrated in the S2 Instrument Processing Facility (IPF) module where it is used for radiometric corrections and resampling.
In the latest releases (from 7.x to 8.0), several features have been added as new SAR sensor models and new applications, and the OSSIM dependency - used for geometric sensor modelling and metadata parsing – has been removed in favor of functionalities available in GDAL. The aim of the presentation is to present the major features of OTB, the latest updates, the future features and architecture of the library and how OTB is used at CNES and CS Group to process data from scientific and developer points of view.
The Orfeo ToolBox is used as development framework for satellite image processing over large dataset in several operational projects. Indeed, its image processing functionalities (multithreading, streaming, ram configuration) allow to process big images quite fast. The operational processing chains use OTB from the Python API and C++ API.
Among the optical chain processing using OTB, we can list: MAJA, WASP, BIOPHY and IOTA2. MAJA (Maccs-Atcor Joint Algorithm) is an atmospheric correction and cloud screening software, based on multi temporal and multi-spectral processing. This chain uses L1C products to generate high quality L2A surface reflectance time series for Landsat8, Venus, and Sentinel 2 missions, it is mainly used by THEIA distribution center. The core algorithms of Maja are based on the Orfeo Toolbox. To process a product, the chain uses aerosol contents, cloud and shadow detection and various atmospheric effects to estimate accurate surface reflectance values. The main problem of the L2A products is the presence of clouds in time series which is why WASP was created. Indeed WASP (Weighted Average Synthesis Processor) delivers L3 products which provide monthly syntheses of cloud-free reflectance for Sentinel2 and Venus L2A products distributed by THEIA. This processor mainly includes a directional correction to normalize data and a weighted average of surface reflectance. Two other operational chains which uses OTB are BIOPHY - the goal of this processing chain is to create L2A products containing biophysical variables (FAPAR, FCOVER, LAI) related to the presence of vegetation in the image over a year – and IOTA2 - a soil occupation processor over a year of Sentinel1 and Sentinel2 data, the algorithms use the classification toolbox provided by OTB to process large areas, to determine the areas covered by buildings.
OrfeoToolBox is also used for radar processing chain, like diapOTB or S1Tiling. S1TIling is a generic processing chain for Sentinel-1 time series developed with open-source software. Its main goal is to produce time series of Analysis ready data S1 images for large areas. The algorithms are using the SAR processing toolbox from OTB to take profit of its in-memory pipelining capabilities. DiapOTB is a differential SAR interferometry processing. It uses two SAR images of the same portion of the Earth’s surface taken at different time as input and aims to analyze potential events (earthquake, destruction …) by highlighting differences between SAR images.
To conclude, OTB is the central framework for a large scale of operational chains in remote sensing. Its genericity permits to cover a lot of use cases in one single tool. Note that it is also distributed in other projects like WorldCereal, SNAP, AI4GEO as toolbox or RUS as training and formation aim.