This is a collection of libraries I initially wrote when I was consulting. I wanted a common platform of unencumbered code for use with different clients, so I wouldn't have to start from scratch each time. I use it as a foundation for much of my software.
The current version number is shown in file VERSION.TXT, and recent changes can be seen in the file RELEASE_NOTES.TXT. I've released it under a pretty permissive license (LGPL) already, but let me know if you want different terms.
These libraries are in regular use under Linux (64-bit). They should be portable to any platform supporting C++11, but they haven't been tested under windows in a very long time.
There are several libraries in this collection, with the higher-level libraries building on the lower-level ones. Here is a quick ascii table illustrating the dependencies between the libraries. To save columns, I've omitted the "brick" prefix in library names, and used three-letter abbreviations in the column headers. The final rows/columns of this table capture dependencies on external libraries (LAPACK/BLAS and libpng/libpng++). We're currently using CMake to build, and it should track these dependencies for you pretty painlessly.
Dependencies When Not Building Unit Tests
-----------------------------------------------------------------------------
Library | Depends on |
|CMN|PTB|TST|NUM|LAL|RND|OPT|UTL|GEO|CTV|PXG|MTF|LPK|PNG|
Common (CMN) | X | | | | | | | | | | | | | |
Portability (PTB) | X | X | | | | | | | | | | | | |
Test (TST) | X | | X | | | | | | | | | | | |
Numeric (NUM) | X | | | X | | | | | | | | | | |
LinearAlgebra (LAL) | X | | | X | X | | | | | | | | X | |
Random (RND) | X | X | | | | X | | | | | | | X | |
Optimization (OPT) | X | | | X | X | | X | | | | | | X | |
Utilities (UTL) | X | X | | | | | | X | | | | | | |
Geometry (GEO) | X | | | X | X | | | | X | | | | X | |
ComputerVision (CTV) | X | | | X | X | X | X | | X | X | | | X | X |
PixelGraphics (PXG) | X | | | X | | | | | X | | X | | | |
Iso12233 (MTF) | X | | | X | X | X | X | | X | X | | X | X | X |
LAPACK (LPK) | | | | | | | | | | | | | | |
libPng++ (PNG) | | | | | | | | | | | | | | |
-----------------------------------------------------------------------------
Each library comes with a bunch of unit tests. If you want to build the unit tests, the dependencies get a little deeper, as the tests often depend on libraries that the tested code doesn't. Again, CMake should invisibly track this for you.
Dependencies When Building Unit Tests
-----------------------------------------------------------------------------
Library | Depends on |
|CMN|PTB|TST|NUM|LAL|RND|OPT|UTL|GEO|CTV|PXG|MTF|LPK|PNG|
Common (CMN) | X | | | | | | | | | | | | | |
Portability (PTB) | X | X | | | | | | | | | | | | |
Test (TST) | X | | X | | | | | | | | | | | |
Numeric (NUM) | X | X | X | X | | | | | | | | | | |
LinearAlgebra (LAL) | X | | X | X | X | | | | | | | | X | |
Random (RND) | X | X | | | | X | | | | | | | X | |
Optimization (OPT) | X | | | X | X | | X | | | | | | X | |
Utilities (UTL) | X | X | X | | | | | X | | | | | | |
Geometry (GEO) | X | | X | X | X | | X | X | X | | | | X | |
ComputerVision (CTV) | X | X | X | X | X | X | X | X | X | X | | | X | X |
PixelGraphics (PXG) | X | | | X | | | | | X | | X | | | |
Iso12233 (MTF) | X | X | X | X | X | X | X | X | X | X | | X | X | X |
LAPACK (LPK) | | | | | | | | | | | | | | |
libPng++ (PNG) | | | | | | | | | | | | | | |
-----------------------------------------------------------------------------
We group these libraries into three subgoups, as described in the following subsections.
The libraries listed in this section have been around for a while. They are well tested, and their interfaces are stable. Features get added from time to time, but we're generally careful not to break existing user code.
This library is the foundation for all of the other libraries in the brick collection. It provides classes for working with exceptions, reference counting (now largely obsoleted by shared pointers), portable numeric types with known precision (the integer types are now obsoleted by standard sized types, but the floating point types are still relevant), byte order, and more.
This library was created to encapsulate whatever platform-specific code is required by other, higher level libraries in the brick suite. Currently it contains some functions for dealing with filesystem operations, some time-related functions, and an incomplete implementation of snprintf. For the most part, user code should not access routines in brick::portability directly. Most of this functionality is already exposed through higher level libraries like brickUtilities. As with brickCommon, updates to the standard have made a bunch of this code obsolete.
This is a unit testing library, similar to CPPUNIT or boost::test, but less featureful. It provides classes and macros for writing tests and coordinating their execution. Currently doesn't include any facility for interrupting tests that run too long, and lacks a formal schema for test output, but it runs the brick tests without requiring any outside dependencies.
This is the original library in the collection. It has code for 1D, 2D, 3D, and ND arrays, coordinate transformations, and much more. This is a good library to use if you need to port Numeric/numpy code from Python to C++. Some recent additions to the library aren't stable, but they're marked as such.
This library provides C++ wrappers around LAPACK linear algebra routines. Currently it includes routines for matrix inverse, solution of linear equations, SVD, determinant, eigenvectors/eigenvalues of symmetric real matrices, etc. A few routines are now templatized to work with arbitrary types, but obviously you won't get the LAPACK optimizations with types that aren't supported by LAPACK.
This library provides a C++ wrapper around the LAPACK pseudo-random number generator. It is is now largely obsoleted by updates to the C++ standard library (see standard header ).
This library provides classes implementing Levenberg-Marquardt nonlinear least squares optimization, Quasi-Newton optimization, and Downhill Simplex nonlinear optimization algorithms. These classes follow the algorithms described in Numerical Recipes in C, by Press and Flannery. This has been a stable library for some time, but we may extend it soon to take advantage of sparse matrices. If this happens, there may be some non-backwards compatible interface changes. Note that the author now uses the Ceres nonlinear least squares solver (ceres-solver.org) for most nonlinear least squares optimization problems.
This library provides some general routines & classes for manipulating strings, parsing commandline arguments, doing filesystem operations, working with dates & times, and a few other odds and ends. It started out as a C++ replacement for the python modules string and path. It is now gradually being replaced by updates to the C++ standard library.
The libraries listed in this section are still growing. Please feel free to use them, but bear in mind that they may change out from under you.
This library provides some general routines for dealing with 2D and 3D geometry. They originally lived in brickComputerVision (back when brickComputerVision was called dlrComputerVision), but were broken out into a separate library in hopes that they have value of their own. It is still in the very early stages of development.
This library includes for basic morphological operations, edge detection, simple deformable contours, histogram equalization, camera models, and a continually growing list of computer vision related things. It's hard to compete with OpenCV these days, but the code in this library is often more readable.
This library provides some classes and functions for drawing in images and image-like objects. It is still in early development.
This library implements the e-SFR algorithm described in ISO 12233:2017. Use it if you need to compute the MTF of an optical system using a slanted edge target. The code isn't super-robust yet (as of May, 2019), but seems to be doing the right things. It passes its unit tests...
This library is intended to provide some classes and functions for working with sparse matrices, but it hasn't been written yet.
Bugfixes and patches welcome! Please see the file LICENSE.TXT in this directory for up-to-date contact information.
Please see the file LICENSE.TXT in this directory.
The distribution includes CMakeLists.txt files to allow building with CMake. The traditional way to build on a linux machine is to start in the top level of the source tree and issue these commands:
> mkdir build
> cd build
> cmake -DCMAKE_INSTALL_PREFIX=/foo/bar ..
> make
> make test
> make install
Thanks, David LaRose