You will find here a short description of the main characteristics of the Orfeo ToolBox software, as well as a list of available functions and algorithms.

Main characteristics

	The power of many – We always look for competitive third parties before writing code. Orfeo ToolBox is built on top of ITK, a popular C++ library for the processing of medical images, and relies on many open source software to implement its functions : GDAL for reading/writing raster/vector data, OSSIM for sensor modeling, OpenCV for machine learning…
	Built for all – Orfeo ToolBox source code is written in a portable way so that OTB supports all major operating systems. Binary packages are provided for many of them. Have a look to the Download page.
	Scalable – Orfeo ToolBox has been originally developed for Very High Resolution imagery (VHR), and as such is able to scale up to the size of those images. Most of the algorithms implemented in OTB supports piece-wise processing, allowing to process very large image under memory constraints. Most of them are also seamlessly multi-threaded.
	Flexible – Orfeo ToolBox C++ API is a collection of algorithmic components that spawn almost unlimited combination for processing chains. Even when using the applications which are tools providing processing chains for standard remote sensing tasks without writing one line of code, flexibility is not far. Found that the application is not exactly what you need ? Fork it and write your own application!
	State of the art – We try our best to keep Orfeo ToolBox with the state of the art of remote sensing image processing, while increasing the number of functions available. OTB can perform in many domain : geometric and radiometric preprocessing,image classification, fusion, change detection, object based image analysis, feature extraction…

Functions and algorithms

Supported formats Image formats supported by GDAL (tiff, hrd, img, data providers specifics …) Vector formats supported by GDAL/OGR (shapefile, sqlite, postgis …) Metadata formats supported by OSSIM and OSSIM plugins (Pléiades and other ADS sensors, DigitalGlobe sensors, NITF) Digital Elevation Models supported by OSSIM (SRTM, DTED and generic geotiff DEM) KMZ image export	Image manipulation Formats and encoding conversion Bands or Region of Interest extraction, bands stacking or splitting, image mosaicking Linear and non-linear filtering Morphological pyramid and morphological profiles Fourier and Wavelets transforms Generic band calculus Color-mapping Block-matching Point sets density estimation filters	Data pre-processing Radiometric calibration and atmospheric corrections (TOA, TOC) Geometric modeling provided by OSSIM Accuracy refinement based on ground control points Registration (sensor to sensor) Ortho-rectification (sensor to ground) Pan-sharpening Stereo-rectification (pseudo-epipolar resampling)
Features extraction SIFT and SURF keypoints detection Keypoints matching Line Segment Detector Radiometric indices Spectral angle Local statistics Local Geometric Hu and Flusser moments Haralick, PanTex and Structural Feature Set texture features Local Histogram of Oriented Gradients	Image Segmentation Conversion between labeled raster and GIS vector file Watershed, connected-components, MeanShift with raster or vector results Large Scale MeanShift framework for segmentation of very large images Object Based Image Analysis with attributes computation and filtering	Image Classification Supervised image classifier training with 9 different machine learning algorithms, including SVM and Random Forest Multi-threaded image classification filter Various unsupervised (clustering) algorithms such as K-Means or Self Organizing Maps Confusion matrix computation with respect to reference or ground truth data Classification map regularization Fusion of several classification maps Object Based Image Classification
Change detection Local metric based change detection framework Multi-variate Alteration Detector Change-detection by supervised classification	Hyperspectral processing Dimensionality reduction with PCA, NAPCA, ICA, MAF Dimensionality estimation with Virtual Dimensionality and ELM VCA endmembers extraction Unmixing with ISRAU, MDMD-NMF, NCLSU, sparsity based algorithms Local RX anomaly detector	SAR processing SAR image manipulation (amplitude and phase image filters) TerraSAR-X radiometric calibration Speckle reduction filters (Frost, Lee) SAR specific features extraction algorithm (Touzi detector, line Correlation and line ratio detector) Polarimetric analysis and synthesis filters Interferometry