gerhobbelt / olena Goto Github PK

Copyright (C) 2008, 2009, 2010, 2011, 2012, 2014 EPITA Research and
Development Laboratory (LRDE).

This file is part of Olena.

Olena is free software: you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation, version 2 of the License.

Olena is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with Olena.  If not, see <http://www.gnu.org/licenses/>.

The complete GNU General Public License Notice can also be found in
the 'COPYING' file in the root directory.


=====================
Introduction to Olena
=====================

Olena_, a platform dedicated to image processing.

.. _Olena: http://olena.lrde.epita.fr


--------
Overview
--------

Olena is a platform dedicated to image processing.  At the moment it is
mainly composed of a C++ library: Milena.  This library features many
tools to easily perform image processing tasks.  Its main
characteristic is its genericity: it allows to write an algorithm once
and run it over many kinds of images (gray scale, color, 1D, 2D, 3D,
...).

Olena is a project developed by the `EPITA Research and Development
Laboratory (LRDE)`__ since 1997.  We did numerous prototypes and
throwaway experiments before settling into the kind of programming
paradigm which is finally here.

__ http://www.lrde.epita.fr

We do our image processing research using this library, but most
importantly we have gathered (and still do) generic programming
expertise from the library development.

Yet, Olena is an ongoing development project.  Few algorithms or
definitions may change in a near future.  A list of potential changes
is maintained here :
http://olena.lrde.epita.fr/FeaturesSubjectToChange.

Likewise, the documentation does not cover the whole project yet.  In the
`doc/' directory you will find the first draft of a reference
manual.  It includes a quick reference guide and a tutorial.  This is a
good start.  In `milena/tools' and `milena/doc/examples' few sample
programs are available.  Most of them are already used to illustrate
the tutorial.  For the rest, we're afraid you will have to dig the
code or e-mail us.

   Please direct any question or comments to <[email protected]>, or
<[email protected]>.

   Olena also has a web page, located at
<http://olena.lrde.epita.fr>.


-----------------
Required Software
-----------------

Here is a non-exhaustive list of required software required to build
Olena successfully.

   * to compile the user examples:

        - a POSIX shell, like Bash

        - a decent C++ compiler, like GNU C++

        - a `make' utility, like GNU `make'

Optional:

   * to use various image types:

        - Magick++ (either provided by ImageMagick or GraphicsMagick)

        - libtiff

        - GDCM

   * To use graphical user interfaces with Scribo:

	- Qt


   * To recognize text with Scribo:

	- Tesseract


==================
Quick Start Manual
==================

This section summarizes the installation procedure.  For more
information about building and installing Olena, see the next
sections.

To install Olena on your system, create a `_build' directory (even
though it is not mandatory) and type in the classical sequence at the
command prompt::

        mkdir _build
	cd _build
	../configure
	make
	make install (as root)

Note that an installation is specific to the compiler used to install
it.  Indeed, the call to ``../configure`` enables some workarounds
and, consequently, users must compile with the same compiler to avoid
compatibility problems.

Between ``make`` and ``make install``, you may also want to run::

	make check

 ``make check`` will run the test suite to check the whole library.
Running the test suite may require up several hours.


=====================
Detailed Instructions
=====================

-------------
Configuration
-------------

In order to prepare the build process, you need to configure the source
tree.

   Assuming your Olena distribution is uncompressed in directory
`olena-2.1', follow these steps:

     % cd olena-2.1
     % mkdir _build
     % cd _build
     % ../configure

   The build process can be altered by a number of options you can pass
to the `configure' script.  The following sections describe them.

   Additionally, if you are an Olena maintainer (a person who runs
`make distcheck'), _prefer setting `CXXFLAGS' as an environment
variable_: the flags given on the command line to `configure' are not
propagated to recursive runs by `make distcheck'.  Or better: use the
environment CONFIG_SITE to set up a configuration environment (see
Autoconf's manual).


Verbose Display
===============

A more verbose display can be turned on by passing the
`--enable-verbose' flag to configure:

     % ../configure --enable-verbose

Before configuring files, configure will display a summary of the
configuration step.


Installation Path
=================

By default, Olena is installed in the standard "local" directory of
your system.  This is usually `/usr/local' under Unix.

   You can change this path with the following flag:

      --prefix=<installation prefix>


Compiler Selection and Compilation Flags
========================================

By default, `configure' will try to use the first C++ compiler it
encounters on your system.  If `CXX' is not set, it will look, in order,
for:

   - the value of the `CXX' environment variable,

   - the GNU C++ compiler (`g++'),

   - the `c++' or `gpp' commands on your system,

   - `aCC', the HP-UX standard C++ compiler,

   - the `CC', `cxx', `cc++' or `cl' commands on your system,

   - KAI's C++ compiler (`KCC'),

   - `RCC', `xlC_r' or `xlC'.

   You can override the detection system by passing your favorite
compiler name to `configure', as follows:

     % ../configure CXX=<your-favorite-C++-compiler>

   As an alternative, you can also set the environment variable `CXX'.


   For some compilers (GNU g++ and Intel's icpc to some extent) ,
`configure' will use default CXXFLAGS.  You can override the default
C++ flags by giving `configure' your selection of flags:

     % ../configure CXXFLAGS="<your-favorite-flags>"


Additional Components
=====================

In additional to Milena, several build targets can be enabled.  These
parts are called "components", and you can obtain a list of them by
running:

     % ../configure --help

Scribo
------

Scribo is an optional component of Olena providing Document Image
Analysis tools and routines. It is based on Milena.

To enable the installation of this module use::

       ./configure --enable-scribo

Swilena
-------

Swilena is an optional component of Olena exposing Milena to other
languages thanks to the Simplified Wrapper and Interface Generator
(SWIG_).

.. _SWIG: http://www.swig.org

For the moment, only some Python_ bindings are provided.  They are
disabled by default because they require extra dependencies (SWIG and
Python).

.. _Python: http://www.python.org

To enable the installation of this module use::

       ./configure --enable-swilena

Tools
-----

Sample tools are shipped with the tarball.  To enable the installation of
these tools use::

       ./configure --enable-tools

Applications
------------
Sample applications are shipped with the tarball.  To enable the
installation of these applications use::

       ./configure --enable-apps

Input/output libraries
----------------------

To read/write TIFF images with Olena, libtiff is required.  If
``configure`` is unable to find libtiff on your system, you can help
it by specifying the base directory of libtiff, e.g.::

       ./configure --with-tiff=/usr/local

To read/write DICOM images with Olena, GDCM is required.  Likewise,
you can tell ``configure`` where to find it by giving its install
prefix, e.g.::

       ./configure --with-gdcm=/usr/local

Olena uses the Magick++ API provided by ImageMagick or GraphicsMagick
to read and write images in common formats.  As for other optional
dependencies, you can specify where such a package is located (if
needed) at configuration time, using either::

       ./configure --with-imagemagickxx=/usr/local/

or

       ./configure --with-graphicsmagickxx=/usr/local/

respectively.

If ``configure`` finds both ImageMagick and GraphicsMagick in your
system (either automatically or because you used one of the above
flags), the former is chosen by default.  If you want to select
GraphicsMagick instead, tell configure explicitly that you do not want
to use ImageMagick.  For instance, in the case where configure is able
to discover GraphicsMagick by itself, the following command line is
enough::

       ./configure --with-imagemagickxx=no

Otherwise, it suffices to combine it with the option telling where
GraphicsMagick is located::

       ./configure --with-imagemagickxx=no --with-graphicsmagickxx=/usr/local

Other libraries
---------------

The Boost Tuple library is used to implement a tuple accumulator
class.  This project is a part of the Boost libraries; you can help
configure find these libraries using the `--with-boost flag':

       ./configure --with-boost=/usr/local/


--------
Building
--------

Once your build directory is `configure'd, you can run

     % make

to recursively build all the selected components.


   Additionally, you can build and run the test suite with:

     % make check

However, this process is time- and memory- consuming, and you probably
do not need it except if you are developing/debugging Olena.


----------
Installing
----------

To install the Olena headers and additional files on your system, run:

      % make install

from the build directory.

   If not overridden with `--prefix', this will install:

   * the headers in `/usr/local/include/mln/',

   * some applications and tools in `/usr/local/bin/',

   * sample images and meshes in `/usr/local/share/olena/images/',

   * the documentation in `/usr/local/share/doc/olena/`

And optionally:

   * Swilena's Python bindings in `/usr/local/lib/python2.x/site-packages/',


   You can later remove Olena from your system by running

      % make uninstall

from the build directory (if you have kept it).  We recommend the use
of GNU Stow (or any similar program) during the installation of Olena,
to make the uninstallation of Olena easier.


=====================
Layout of the Tarball
=====================

The Olena project directory layout is as follows:

build-aux
   Auxiliary tools used by the GNU Build System during ``configure``
   and ``make`` stages.

external
   Sources of Shipped dependencies.

m4
   Extra Autoconf macros.

milena

  apps
     Application examples.

  mln
     Headers of the Milena library.

  tests
     The test suite.

  doc
     The documentation.

  tools
     Example tools.

  mesh
     Some 3D meshes, mostly used for test purpose.

  img
     Some (2D) images, mostly used for test purpose.

  demos
     Demos of Milena.

scribo

  demo
     Full applications using Scribo DIA capabilities.

  scribo
     Headers of the Scribo framework.

  src
     Example tools.

swilena

  python
     Some Python bindings for Milena.


===================
Supported Platforms
===================

Olena has been tested on the following configurations:

===========================  =============================================
System                       Compiler
===========================  =============================================
GNU/Linux on x86-64          g++ (GNU GCC) 4.7, 4.8
GNU/Linux on x86-64          clang++ (LLVM) 3.5
OS X (10.8) on Intel 64      g++ (GNU GCC) 4.8
OS X (10.8) on Intel 64      clang++ (Apple LLVM) 5.0
===========================  =============================================


According to the wanted features, some dependencies may be required:


To enable I/O with TIFF images, `libtiff`_ must be installed.

.. _libtiff: http://www.libtiff.org/

To enable I/O with GDCM images, `GDCM`_ must be installed.

.. _GDCM: http://sourceforge.net/apps/mediawiki/gdcm/

To support many image formats in Olena's I/O system, `Magick++`_ must
be installed.

.. _Magick++: http://www.imagemagick.org/Magick++/

The Boost Tuple Library from the `Boost`_ Project is needed if you
want to support tuple accumulators.

.. _Boost: http://www.boost.org/

Apart from GDCM, these dependencies are commonly provided by the
package management systems (e.g., Debian's APT, Mac OS X's Mac Ports,
etc.).  We recommend using package managers instead of installing
dependencies by hand.


See Also
========

There are other sources of interest in the distribution.

- Headline news about the project can be found in the file ``NEWS`` at
  the root of the source tree.

- The library reference HTML documentation, generated by Doxygen_, is
  located in ``doc/user/html/``.

.. _Doxygen: http://www.doxygen.org


License
=======

Olena is released under the GNU General Public Licence.  See the file
``COPYING`` (at the root of the source tree) for details.


Contacts
========

The team can be reached by mail at [email protected].  The snail
mail address follows.

* Olena - LRDE

  | Laboratoire de Recherche et Développement de l'EPITA (LRDE)
  | 14-16 rue Voltaire
  | FR-94276 Le Kremlin-Bicêtre CEDEX
  | France


Bibliography
============

Further information about Olena and image processing can be found in
the following papers.

About Image Processing Programming:

   * `Writing Reusable Digital Topology Algorithms in a Generic Image
     Processing Framework`.  Roland Levillain, Thierry Géraud, Laurent
     Najman.  In Ullrich Köthe, Annick Montanvert, Pierre Soille
     (editors), WADGMM 2010 -- Proceedings of the Workshop on
     Applications of Digital Geometry and Mathematical Morphology),
     pages 140-153, Lecture Notes in Computer Science, volume 7346,
     Springer-Verlag Berlin Heidelberg, 2012.

   * `Une approche générique du logiciel pour le traitement d'images
     préservant les performances`.  Roland Levillain, Thierry Géraud,
     Laurent Najman.  In the proceedings of the 23rd Symposium on
     Signal and Image Processing (GRETSI)
     http://www.gretsi2011.org/
     Bordeaux, France, September 5 - 8, 2011.

   * `Why and How to Design a Generic and Efficient Image Processing
     Framework: The Case of the Milena Library`.  Roland Levillain,
     Thierry Géraud, Laurent Najman.  In the proceedings of the 2010
     International Conference on Image Processing (ICIP)
     http://www.icip2010.org/
     Hong Kong, September 26 - 29, 2010.

   * `Milena: Write Generic Morphological Algorithms Once, Run on Many
     Kinds of Images`.  Roland Levillain, Thierry Géraud, Laurent
     Najman.  In the proceedings of the 9th International Symposium on
     Mathematical Morphology (ISMM)
     http://www.cs.rug.nl/~ISMM09/
     Groningen, The Netherlands, August 24 - 27, 2009.

   * `Generic Algorithmic Blocks Dedicated to Image Processing`,
     Jérôme Darbon, Thierry Géraud, Patrick Bellot.  In the proceedings
     of ECOOP PHD Oslo, Norway, June 2004.

   * `Generic Implementation of Morphological Image Operators`, Jérôme
     Darbon, Thierry Géraud, and Alexandre Duret-Lutz.  In the
     proceedings of the International Symposium On Mathematical
     Morphology VI (ISMM 2002), April 3-5, 2002, Sydney, Australia.

   * `Applying Generic Programming to Image Processing`, Thierry
     Géraud, Yoann Fabre, and Alexandre Duret-Lutz.  In the
     proceedings of the IASTED International Conference on Applied
     Informatics (AI'2001) - Symposium Advances in Computer
     Applications, ACTA Press, pages 577-581, Innsbruck, Austria,
     February 2001.

   * `Obtaining Genericity for Image Processing and Pattern
     Recognition Algorithms`.  Thierry Géraud, Yoann Fabre, Alexandre
     Duret-Lutz, Dimitri Papadopoulos-Orfanos, and Jean-François
     Mangin.  In the proceedings of the 15th International Conference
     on Pattern Recognition (ICPR'2000), IEEE Computer Society,
     vol. 4, pages 816-819, Barcelona, Spain, September 2000.

About Generic Programming Paradigm:

   * `Towards a Software Architecture for Generic Image Processing`.
     Roland Levillain.  Ph.D. thesis, Universit\'e Paris-Est,
     Marne-la-Vallée, France, 2011.

   * `Semantics-Driven Genericity: A Sequel to the Static C++
     Object-Oriented Programming Paradigm (SCOOP 2)`.  Thierry Géraud,
     Roland Levillain.  In the proceedings of the 6th International
     Workshop on Multiparadigm Programming with Object-Oriented
     Languages
     http://homepages.fh-regensburg.de/~mpool/mpool08/welcome.html
     Paphos, Cyprus July 7, 2008.

   * `Static C++ Object-Oriented Programming (SCOOP)`, Nicolas Burrus,
     Alexandre Duret-Lutz, Thierry Géraud, David Lesage, and Raphaël
     Poss.  In the proceedings of the Workshop on Multiple Paradigm
     with OO Languages (MPOOL'03), Anaheim, CA, October 2003.

   * `Generic Design Patterns in C++`.  Alexandre Duret-Lutz, Thierry
     Géraud, and Akim Demaille.  In the proceedings of the 6th USENIX
     Conference on Object-Oriented Technologies and Systems
     (COOTS'2001), pages 189-202, San Antonio, Texas, USA,
     January-February 2001.

   * `Olena: a Component-Based Platform for Image Processing, mixing
     Generic, Generative and OO Programming`.  Alexandre Duret-Lutz.
     In the proceedings of the 2nd International Symposium on
     Generative and Component-Based Software Engineering (GCSE 2000),
     Young Researchers Workshop (published in "Net.ObjectDays2000";
     ISBN 3-89683-932-2), pages 653-659, Erfurt, Germany, October
     2000.


   You can download these papers and related material from
<http://publications.lrde.epita.fr>.



.. Local Variables:
.. mode: rst
.. End: