The DIPlib project contains:
-
DIPlib, a C++ library for quantitative image analysis. It has been in development at Delft University of Technology in The Netherlands since 1995. The 3.0 release of DIPlib represents a complete rewrite in modern C++ of the library infrastructure, with most of the image processing and analysis algorithms ported from the previous version, and some of them improved significantly. Many of the improvements to DIPlib are a result of porting over concepts and ideas first implemented in DIPimage. Read the DIPlib documentation.
-
DIPimage, a MATLAB toolbox for quantitative image analysis. It has been in development at Delft University of Technology in The Netherlands since 1999. Read the DIPimage User Manual.
-
PyDIP, Python bindings to DIPlib. This is currently a thin wrapper that exposes the C++ functionality with little change. the PyDIP User Manual
-
DIPviewer, an interactive image display utility. It is usable from C++, Python and MATLAB programs. Within DIPimage this is an optional alternative to the default MATLAB-native interactive display utility. Read the DIPviewer documentation.
-
DIPjavaio, an interface to OME Bio-Formats, a Java-based library that reads hundreds of image file formats. This module is usable from C++ and Python. DIPimage has a different way of interfacing with Bio-Formats. Read the DIPjavaio documentation
See the DIPlib website for more information.
To build the library you will need a C++14 compliant compiler and CMake. Use the following commands to build:
mkdir target
cd target
cmake /path/to/dip/root/directory
make -j install
For detailed instructions, see INSTALL.md
.
Unless you want to use Cygwin or MinGW (see above), we recommend Microsoft Visual Studio 2019. You'll also need CMake.
Using CMake-gui, choose where the source directory is and where to build the binaries. Then press "Configure" and select Visual Studio. Finally, press "Generate". You should now have a Visual Studio solution file that you can open in Visual Studio and build as usual.
For step-by-step instructions, see INSTALL_Windows.md
. See also
INSTALL.md
for additional information and optional dependencies you might want
to install.
When using CMake, and importing the DIP
target into your project in the right way, you will just need
to link against the DIP
target and everything will be configured correctly. Otherwise, there are several
macros that you should define when building any program that links against DIPlib:
If DIPlib was build with the DIP_SHARED_LIBRARY
flag not set, then you need to define the DIP_CONFIG_DIP_IS_STATIC
macro when compiling the code that links against it. Likewise, if the DIP_ALWAYS_128_PRNG
flag was set,
then you must define a DIP_CONFIG_ALWAYS_128_PRNG
macro when compiling your program. Mismatching this flag
could cause your program to not link, or worse, crash at runtime.
The following flags do not need to be matched, but they should be if you want the inline functions to behave the same as the pre-compiled ones:
- flag:
DIP_ENABLE_STACK_TRACE
-- macro:DIP_CONFIG_ENABLE_STACK_TRACE
- flag:
DIP_ENABLE_ASSERT
-- macro:DIP_CONFIG_ENABLE_ASSERT
Also, if your compiler supports __PRETTY_FUNCTION__
, set the macro DIP_CONFIG_HAS_PRETTY_FUNCTION
to
get better stack traces.
For DIPviewer, if DIP_SHARED_LIBRARY
was not set, define the DIP_CONFIG_DIPVIEWER_IS_STATIC
macro.
Also define DIP_CONFIG_HAS_FREEGLUT
or DIP_CONFIG_HAS_GLFW
depending on which back-end is used.
For DIPjavaio, if DIP_SHARED_LIBRARY
was not set, define the DIP_CONFIG_DIPJAVAIO_IS_STATIC
macro.
Also define DIP_CONFIG_HAS_DIPJAVAIO
.
We use the issue tracker on GitHub to manage bug reports.
See CONTRIBUTING.md
for instructions on how to report a bug.
Feel free to submit a pull request on GitHub. Please follow
our style guide and make sure to read
CONTRIBUTING.md
first.
Documentation can always be improved (also, PyDIP has hardy any at all!). If you want to help write documentation,
or create tutorials for how to use the library, read CONTRIBUTING.md
, then submit a
pull request on GitHub.
If you found an error in the documentation, we consider this a bug. See above how to report it.
Copyright 2014-2021 Cris Luengo and contributors
Copyright 1995-2014 Delft University of Technology
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this library except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
(or see the LICENSE.txt
file in this distribution)
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
DIPlib includes the following libraries / external code:
-
Eigen 3
Copyright 2008 Gael Guennebaud
Copyright 2007-2011 Benoit Jacob
Mozilla Public License Version 2.0
seedependencies/eigen3/LICENSE.txt
(we do not use any of the components with more restrictive licenses) -
DocTest (its use can be disabled)
Copyright 2016-2019 Viktor Kirilov
MIT License
seedependencies/doctest/LICENSE.txt
-
The DFT algorithm out of OpenCV 3.1
Copyright 2000 Intel Corporation
Intel License Agreement For Open Source Computer Vision Library
seesrc/transform/opencv_dxt.cpp
-
PCG Random Number Generation for C++
Copyright 2014-2017 Melissa O'Neill and the PCG Project contributors
Apache 2.0 License, or The MIT License, at your option
seeinclude/diplib/private/pcg_*.hpp
-
robin-map, a fast hash map and hash set
Copyright 2017 Thibaut Goetghebuer-Planchon
MIT License
seeinclude/diplib/private/robin_*.h
-
A fast 2D labeling algorithm out of YACCLAB
Copyright 2016-2017 Costantino Grana, Federico Bolelli, Lorenzo Baraldi and Roberto Vezzani
3-Clause BSD License
seesrc/regions/labelingGrana2016.h
-
A few color maps from colorcet
Copyright 2017 Peter Kovesi
1-Clause BSD-like License
seesrc/display/colormap.cpp
(about half-way down the file) -
libics (its use can be disabled)
Copyright 2015-2017 Scientific Volume Imaging Holding B.V.
Copyright 2000-2013 Cris Luengo and others
GNU Lesser General Public License, Version 2.1
seedependencies/libics/GNU_LICENSE
-
LibTIFF (its use can be disabled)
Copyright 1988-1997 Sam Leffler
Copyright 1991-1997 Silicon Graphics, Inc.
2-Clause BSD-like Licence
seedependencies/libtiff/COPYRIGHT
-
zlib (used by libics and LibTIFF, its use can be disabled)
Copyright 1995-2017 Jean-loup Gailly and Mark Adler
zlib Licence
seedependencies/zlib/LICENSE.txt
-
libjpeg (its use can be disabled)
Copyright 1991-2018, Thomas G. Lane, Guido Vollbeding
Custom BSD-like Licence
seedependencies/libjpeg/README
-
pybind11 (only used in the Python bindings)
Copyright 2016 Wenzel Jakob
3-Clause BSD License
seedependencies/pybind11/LICENSE
-
python-javabridge (only used in the Python bindings)
Copyright 2003-2009 Massachusetts Institute of Technology
Copyright 2009-2013 Broad Institute
3-Clause BSD License
seepydip/src/loadjvm.py
-
A few MATLAB scripts from OME Bio-Formats (used in DIPimage)
Copyright 2012-2021 Open Microscopy Environment
2-Clause BSD License
seedipimage/private/bf*.m
(the full Bio-Formats library needs to be installed for these to be useful) -
Modified CSS files from dox++, which originally came from m.css (for documentation)
Copyright 2017, 2018, 2019 Vladimír Vondruš
MIT License
Note that all of these have permissive open-source licenses similar in spirit to the Apache License.
DIPlib also optionally links against:
- FFTW3 (as installed on your system, not used by default)
Copyright 2003, 2007-14 Matteo Frigo
Copyright 2003, 2007-14 Massachusetts Institute of Technology
GNU General Public License, Version 2 (not compatible with the Apache License)
DIPviewer links against the following libraries:
-
FreeGLUT (as installed on your system, alternative to GLFW)
Copyright 1999-2000 Pawel W. Olszta
X-Consortium license -
GLFW (as installed on your system, alternative to FreeGLUT)
Copyright 2002-2006 Marcus Geelnard
Copyright 2006-2011 Camilla Berglund
BSD-like license -
OpenGL (as installed on your system)
(free from licensing requirements)
DIPjavaio links against:
-
JDK 8 (as installed on your system)
Copyright, probably Sun and/or Oracle
GNU General Public License, Version 2 (not compatible with the Apache License) -
OME Bio-Formats (optional, as installed on your system)
Copyright 2012-2017 Open Microscopy Environment
GNU General Public License, Version 2 (not compatible with the Apache License)
Note that OME Bio-Formats is optional, but DIPjavaio is currently useless without it. The DIPjavaio sources provided with this project are shared under the same licence as the rest of the project (Apache 2.0), but by linking with Bio-Formats, your whole program will become GPL. This module is not suitable for use in commercial, closed-source software.