Orfeo toolbox

OTB provides:[3]

• Image access: read/write access for most remote sensing image formats (using GDAL), meta-data access, visualization [4]
• Data access: vector data access (shapefile, kml), DEM model, lidar data [5]
• Filtering: blurring, denoising, enhancement for optical or radar data [6]
• Feature extraction: texture computations including Haralick, SFS, Pantex, Edge density, points of interest, alignments, lines, SIFT, SURF [7]
• Image segmentation: region growing, watershed, level sets [8]
• Classification: K-means, SVM, Markov random fields and access to all OpenCV machine learning algorithms[9]
• Change detection [10]
• Stereo reconstruction from images
• Orthorectification and map projections (using ossim) [11]
• Radiometric indices (vegetation, water, soil) [12]
• Object-based segmentation and filtering
• PCA computation
• Visualization: a flexible visualization system, customizable via plugins;
• and more.

OTB is a C++ library, based on Insight toolkit (ITK), a medical image processing library.

Bindings are developed for Python and are available as the separate OTB-Wrapping project. A blog post on the orfeo-toolbox blog details an example using the python wrapping [13]

A method to use OTB components within IDL/ENVI has been published.

One of the OTB user defined a procedure to use the library capabilities from MATLAB.[14]

Since late 2009,[15] some modules are developed as processing plugins[16] for QGIS. Modules for classification, segmentation, hill shading have provided. This effort has not been funded so far and relies only on volunteers.

OTB algorithms are now available in QGIS through the processing framework Sextante.

Additionally to the library, several applications with GUI are distributed. These application enable interactive segmentation, orthorectification, classification, image registration, etc...

The OTB-Applications package makes available a set of simple software tools which were designed to demonstrates what can be done with OTB. Many users started using these applications for real processing tasks, so we tried to make them more generic, more robust and easy to use. It supports raster and vector data and integrates most of the already existing OTB applications. The architecture takes advantage of the streaming and multi-threading capabilities of the OTB pipeline. It also uses cool features such as processing on demand and automagic file format I/O. The application is called Monteverdi,[17] since this is the name of the Orfeo composer. This is also in memory of the great (and once open source) Khoros/Cantata software.[18]

In 2013, Monteverdi software have been revamped to take into account users feedbacks regarding how useful the tool was, but also regarding what should be improved to move toward greater usability and operationnality. Monteverdi concept has been reworked into a brand new software called Monteverdi2, enlightened by this experience.

OTB was initially distributed under the french Open Source license CeCILL (similar and compatible with the GNU GPL) and is now available under the Apache 2.0 license.

The development started in January 2006 [19] with the first release in July 2006.[20] The development version is publicly accessible.[21]

• Official website