BRINGING THE RASTER PROCESSING ALGORITHMS OF THE ORFEO TOOLBOX MONTEVERDI IN QGIS
The Orfeo Toolbox (OTB) is an Open Source Remote Sensing Image Processing software library developed by CNES. Orfeo Toolbox was presented at FOSS4G in 2009. The aim of the toolbox is to gather a large number of state of the art algorithms for building processing chains for satellite images. The Monteverdi integrated application is based on OTB and provides a user-friendly graphical interface to many of the algorithms available in the OTB library. It also aims at providing higher level capabilities by combining several elements of the library.
The modular architecture of the software allows to easily extend the application with custom algorithms (or algorithm combinations) seamlessly integrated into the main application.
It already provides a large set of building blocks for processing chains and is designed to host virtually an infinite number of modules. Typical processing examples include orthorectification and reprojection of satellite images (based on the sensor model or ground control points), supervised and unsupervised classification (SVM, K-Means), interactive segmentation, basic image processing tasks (threshold, filtering, feature extraction etc). The architecture takes advantage of the streaming and multi-threading capabilities of the OTB pipeline and it also uses great features such as processing on demand and automagic file format I/O. This high level of customization promises unlimited combination of process and the perspective to be able to provide answers to a wide panel of remote sensing product users.
Since FOSS4G 2009, simple OTB processing chains have been made available inside QGIS through its plug-in framework. After this proof of concept was made, it seemed straightforward to made the whole Monteverdi application become a QGIS plugin.
This presentation will focus on the latest developments around Monteverdi which allow to use it from within Quantum GIS bringing the power of raster processing algorithms to the user-friendly interface of QGIS.
Emmanuel Christophe - CRISP, National University of Singapore
Manuel Grizonnet - CNES
Julien Michel - CNES
Jordi Inglada - CNES/CESBIO
Tishampati Dhar - Apogee Imaging International