[IPOL discuss] [YAIPL] "population" image processing library

[Jerome Darbon]

Hi,

The authors of Olena are also close to Cachan if you want to meet with 
them: http://www.lrde.epita.fr/cgi-bin/twiki/view/Olena/

Here is a brief description:
"Olena <http://www.lrde.epita.fr/cgi-bin/twiki/view/Olena/Olena> is a 
software platform dedicated to image processing. At the moment it is 
mainly composed of a C++ library: Milena 
<http://www.lrde.epita.fr/cgi-bin/twiki/view/Olena/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 (grey scale, color, 1D, 2D, 3D...). "

best,
jerome



On 11/06/2011 07:02 PM, Nicolas Limare wrote:
> Hi,
>
> YAIPL = Yet Another Image Processing Library...
> I feel like every week I see a new place where someone was not
> satisfied by existing image processing libraries and decide to write
> their own. And this doesn't even ensure that we will eventually get a
> good one, because they always try to solve a different set of problems.
>
> Today, I found "population", which defines itself as "[...] an
> open-source imaging library in C++ for processing, analysing,
> modelling and visualising including more than 200 algorithms designed
> by V. Tariel under the CeCill licence.
>
> ->  http://population.shinoe.org/
>
> «Why a new library
>
> Since the emergence of applications of the image data-structure, many
> libraries dedicated to image processing/analysis have be distributed
> freely ImageJ, OpenCV, Olena, VTK or commercially,  ADCIS, Matlab
> [3]. In my knowledge, no one respects all these criteria :
>
> * ad vitam æternam : now and in the future, it is and it will be easy
>    to install the software in any platforms. Therefore, the code source
>    should be written in a standard language and should restrict the
>    utilization of external libraries,
> * user-friendly : it is easy to design your own algorithm. Therefore,
>    the library should have a good technical documentation, as nice
>    example [4] and the data-structures should be mapped into classes,
> * optimized : the algorithms can be applied on large data set as 3d
>    image. Therefore, following generic programming paradigm, the
>    procedures should beparametrized to do most of the work at compile
>    time and, therefore, to save time at execution time,
> * modularity : an algorithm is built on elementary bricks. Therefore, a
>    algorithm taxonomy should be done for the definition of the elementary
>    bricks,
> * mathematical equivalence : we have a mathematical equivalence between
>    the algorithm definition and the algorithm implementation. In other
>    words,  an algorithm can be applied on a class of data-structure
>    having some fundamental properties allowing the process of versatile
>    data-structures with a single procedure (whatever the data-structures
>    with some properties, I can operate that.
>
> To fulfil all criteria, I design my home-made library on C++ language,
> called Population.»
>
> The author, Vincent Tariel, is PhD student at Polytechnique. So close
> to Cachan, it could be good to meet him to know more.
>
>
>
> _______________________________________________
> discuss mailing list
> discuss at list.ipol.im
> http://tools.ipol.im/mailman/listinfo/discuss
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://tools.ipol.im/mailman/archive/discuss/attachments/20111107/31fc0965/attachment.html>