VIGRA Reference Manual

• Installation
     how to get started
• Tutorial
     first steps with VIGRA
• Data Structures
     multi-dimensional arrays, graphs, numeric types etc.
  • vigra::TinyVector
       fixed-size arrays
  • vigra::MultiArrayView
       multi-dimensional array interface
  • vigra::MultiArray
       multi-dimensional array class that holds the actual memory
  • vigra::NumpyArray and vigra::NumpyAnyArray
       wrappers offering VIGRA's multi-array API for Python arrays
  • vigra::linalg::Matrix
       array specialization for linear algebra
  • Image Containers
       classes to manage multiple images and pyramids
  • vigra::SplineImageView
       on-the-fly interpolation of 2D images and their derivatives
  • Chunked arrays
       big data (potentially larger than RAM) stored as a collection of rectangular blocks
  • vigra::AdjacencyListGraph
       general graph class
  • vigra::GridGraph
       direct and indirect neighbor graphs on arbitrary dimensional grids
  • vigra::BinaryForest
       a collection of binary trees
  • Pixel and Number Types
       rationals, complex numbers, RGB tuples etc.
    • Fixed Size Integer Types
    • vigra::RGBValue
    • Numeric and Promote Traits of RGBValue
    • vigra::TinyVector
    • vigra::TinyVectorView
    • Numeric and Promote Traits of TinyVector
    • vigra::FFTWComplex
    • Numeric and Promote Traits of FFTWComplex
    • vigra::Rational
    • vigra::FixedPoint
    • vigra::FixedPoint16
  • vigra::BucketQueue and vigra::MappedBucketQueue
       efficient priority queues for integer priorities
  • vigra::Any
       typesafe storage of arbitrary values
  • vigra::ArrayVector
       replacement for std::vector with stronger guarantees (deprecated as of C++ 11)
  • vigra::BasicImage
       deprecated type for 2D images
  • vigra::BasicImageView
       deprecated type for 2D images that use external memory
• Image and Array I/O
     read and write data from/to disk
  • Image Import/Export Facilities
       highlevel import/export interface for 2D images
  • Import/export of volume data.
       import/export interface for volume data
  • Import/Export of Images and Arrays in HDF5 Format
       import/export of arbitrary-dimensional arrays in the HDF5 format
  • vigra::ChunkedArrayHDF5
       automated memory management of huge datasets via HDF5
  • Import/export of the TIFF format
       image import/export interface to call libtiff functions directly
• Parallel Processing
     using std::thread
• Image Processing
     array arithmetic, convolution filters, morphology, color conversion, registration etc.
  • vigra::multi_math
       arithmetic and algebraic expressions for multi-dimensional arrays
  • Point operators for multi-dimensional arrays.
       point operators on multi-dimensional arrays
  • Color Space Conversions
       convert between RGB and other color spaces like L*u*v* and Y'PbPr
  • Filters
       Gaussian filters, smoothing and sharpening, gradients etc.
    • Convolution Filters
         arbitrary-dimensional filters in the spatial and Fourier domains
    • Recursive convolution functions
         recursive filters (1st and 2nd order)
    • Resampling Convolution Filters
         filters changing the array size
    • Gabor Filter
         Gabor filter creation and related functionality
    • Tensor Image Processing
         tensor-valued pixel features
    • vigra::Kernel1D and vigra::Kernel2D
         generic discrete convolution kernels
    • One-dimensional and separable convolution functions
         1D convolution and separable filters in 2 dimensions
    • BorderTreatmentMode
  • Geometric Transformations and Resizing
       arbitrary-dimensional resize and interpolation, mirroring, rotation, affine warping
  • vigra::SplineImageView
       on-the-fly interpolation of 2D images and their derivatives
  • Fast Fourier Transform
       fast Fourier transform for arrays of arbitrary dimension
  • Image Correlation
       estimate the correlation between images
  • Image Registration
       transform images into a common coordinate system
  • Non-linear Diffusion and Total Variation
       edge-preserving smoothing and denoising
  • Distance Transform
       distance transforms in arbitrary dimensions
  • Morphological operators for multi-dimensional arrays.
       separable morphology with parabola structuring functions in arbitrary dimensions
  • Basic Morphological Operations
       2D erosion, dilation, and median with disc structuring functions
  • Noise Normalization
       transform intensity-dependent noise into additive Gaussian noise
  • Camera MTF Estimation
       determine the magnitude transfer function (MTF) of a camera using the slanted edge method
• Image Analysis
     from pixels to structured data
  • Feature Accumulators
       computate multi-dimensional statistics for regions or hte entire image
  • Algorithms to Inspect Images and Functors To Inspect Images
       outdated statistical analysis functions
  • vigra::Threshold
       good old thresholding
  • Connected Components Labeling
       connected components labeling
  • Local Minima and Maxima
       Including extremal plateaus larger than 1 pixel
  • Distance Transform
       including vector distance, eccentricity transforms, and skeletons
  • Edge Detection
       edge detectors based on first and second derivatives
  • Corner Detection
       measure the 'cornerness' at each pixel
  • Symmetry Detection
       measure the local symmetry at each pixel
  • Superpixel Creation
       watersheds, SLIC superpixels, and seeded region growing
  • Graph Data Structures and Algorithms
       graph-based segmentation algorithms
• Graph Data Structures and Algorithms
     graph classes and graph-based algorithms
• Machine Learning
     classification algorithms
• Mathematical Tools
     number types, special functions, linear algebra and optimization etc.
  • Numeric and Promotion Traits
       meta-information about arithmetic types
  • Mathematical Constants
       M_PI, M_SQRT2
  • Mathematical Functions
       and functors
  • Number Types
    • vigra::Rational
    • vigra::TinyVector
    • vigra::autodiff::DualVector
    • vigra::FFTWComplex
    • vigra::FixedPoint16
    • vigra::Quaternion
  • Random Number Generation
       Mersenne twister class and random number functors
  • Polynomials and root determination
       Polynomials and root determination
  • vigra::linalg::Matrix
       array specialization for linear algebra
  • Linear Algebra
       matrix algebra, solution of linear systems, eigenvalue calculation etc.
  • Unsupervised Decomposition
       Unsupervised matrix decomposition methods (pLSA)
  • Optimization and Regression
       ordinary and non-negative least squares, ridge regression, least angle regression (LARS and LASSO)
  • Algebraic Concepts
       Requirements for types that implement arithmetic operations
• Utilities
     error handling, speed measurement
  • Error Reporting
       exceptions and assertions
  • Timing macros for runtime measurements
       macros for taking execution speed measurements
  • VIGRA_FINALLY
       emulation of the 'finally' keyword from Python
  • vigra::Any
       typesafe storage of arbitrary values
  • vigra::BucketQueue and vigra::MappedBucketQueue
       efficient priority queues for integer priorities
  • Ranges and Points
       2-D and N-D positions, extents, and boxes
  • vigra::FilterIterator
       skip elements in a range that don't fulfill a predicate
  • vigra::IteratorAdaptor
       quickly create STL-compatible 1D iterator adaptors
  • Tuple Types
       pair, triple, tuple4, tuple5
  • Utilities to manage pixel neighborhoods
       2D neighborhood definitions (deprecated, use vigra::GridGraph instead)
  • Utilities to manage voxel neighborhoods
       3D neighborhood definitions (deprecated, use vigra::GridGraph instead)
  • vigra::ArrayVector
       replacement for std::vector with stronger guarantees (deprecated as of C++ 11)
• Concepts
     generic interface definitions
  • Algebraic Concepts
       requirements for types that implement arithmetic operations
  • Image Iterators
       2D iterator API
  • Multi-dimensional Array Iterators
       multi-dimensional iterator API
  • vigra::MultiArrayNavigator
       navigator utility for multi-dimensional arrays
  • vigra::FunctorTraits
       requirements for functor traits
  • Crack Edge Image
       topologically correct treatment of interpixel contours
  • Data Accessors
       requirements for data accessors (deprecated)
• Examples
     demonstration programs for VIGRA's usage
• Vigra Matlab
     VIGRA Matlab bindings (unsupported)
• vigranumpy
     VIGRA Python bindings
• Credits and Changelog
     who contributed what?

VIGRA - Vision with Generic Algorithms

Version 1.11.1 by Ullrich Köthe

VIGRA is a computer vision library that puts its main emphasis on flexible algorithms, because algorithms represent the principle know-how of this field. The library was consequently built using generic programming as introduced by Stepanov and Musser and exemplified in the C++ Standard Template Library. By writing a few adapters (image iterators and accessors) you can use VIGRA's algorithms on top of your data structures, within your environment. Alternatively, you can also use the data structures provided within VIGRA, which can be easily adapted to a wide range of applications. VIGRA's flexibility comes almost for free: Since the design uses compile-time polymorphism (templates), performance of the compiled program approaches that of a traditional, hand tuned, inflexible, solution.

VIGRA's design is documented in the chapter "Reusable Software in Computer Vision" by Ullrich Köthe, in: B. Jähne, H. Haußecker, P. Geißler: "Handbook on Computer Vision and Applications", volume 3, Academic Press, 1999. If you don't have the book, you may read a draft of this article. A shorter article STL Style Generic Programming with Images describing some of the ideas has appeared in the January 2000 issue of C++ Report Magazine. The most comprehensive source of information is U. Köthe's PhD thesis Generische Programmierung für die Bildverarbeitung, but this is written in German.

VIGRA is subject to this LICENSE.

You can also subscribe to the VIGRA Mailing List to get instant information about new releases, discuss VIGRA's features and development, and ask the experts for help.