September 2013 - Olena

'void mln::io::magick::load(mln::Image<I>&, const std::string&)'
by Robby Oliver 10 Sep '13

10 Sep '13

I have been trying to make this on FreeBSD 8.2 and 9.1 with no luck. I have tried both IM and GM and I have yet to figure out why I am getting this error when it is correct type. Gxx is 4.2.1 Just wondering if anyone else has run into this problem before...If so is their a patch to get around this?

2 1

mln/version.hh now generated in the source tree
by Roland Levillain 10 Sep '13

10 Sep '13

Hi everybody, I've recently made some changes regarding Milena's header mln/version.hh [1]. This file is produced from milena/mln/version.hh.in, and was originally generated in the *build tree* by configure. With the recent changes in `master', this file is now generated in the *source tree*, by make. If your Olena build tree is different from your Olena source tree (i.e. if you've invoked configure from a directory other than the one where it resides), you'll have to manually remove the old generated milena/mln/version.hh file from your build tree. E.g. if you previously configured Olena in the following manner: cd ~/src/olena mkdir _build cd _build ../configure you'll need to remove mln/version.hh from the build dir this way: rm ~/src/olena/_build/milena/mln/version.hh If you configured the source tree to be the build tree (i.e. by running `./configure' from the source tree), you don't need to do anything special. Notes: [1] See http://git.lrde.epita.fr/?p=olena.git;a=commit;h=5158238ced7f9256ca51e68dc4…. -- Roland Levillain EPITA Research and Development Laboratory (LRDE) 14-16, rue Voltaire - FR-94276 Le Kremlin-Bicêtre Cedex - France Phone: +33 1 53 14 59 45 - Fax: +33 1 53 14 59 22 - www.lrde.epita.fr

1 0

rgb8 image in Olena
by Paula Agregán Reboredo 05 Sep '13

05 Sep '13

Hi, I am Paula, image processing student, and I am working with Olena. I have a doubt: I want to create a RGB image and I want to give each pixel a certain color. I do the following: image2d <value::rgb8> imagePrueb(rows,cols); fill(imagePrueb, literal::white); This way, the problem is the whole image has the same color... How can I set the desired color to each pixel? Thank you. Best regards. Paula.

2 1

Démo de dématérialisation
by Guillaume Lazzara 05 Sep '13

05 Sep '13

Salut tout le monde, j'ai voulu tester ce week-end la démo en ligne de dématérialisation ( http://olena.lrde.epita.fr/demos/document_segmentation.php) de document pour récupérer mon image en pdf mais ça a finit en erreur : "PDF output : /tmp/web_app147489430652208889c261d3.66045018.pdf"Ooops! Something went wrong. Please contact us in order to fix that! olena(a)lrde.epita.fr. Thank you! Ca fait la même chose si je lance la démo avec les images proposées sur le site. Je n'ai pas de problème par contre avec la démo sur les documents historiques. Bon courage et à bientôt. :) A+

2 1

SauvolaMs, for ancient Torah Scrolls Binarization ?
by Albert Berman 05 Sep '13

05 Sep '13

Shalom Olena team, At Ben-Gurion University in Israel we evaluating binarization approaches for ancient Torah Scrolls right now. You SauvoulaMS approach :"Efficient Multiscale Sauvola's Binarization". seems to be very accurate even despite of inhomogeneous background! So we would be really gad to use it and see how it performs ahead of others Binarization algorithms. I already installed Olena 2.0 under Ubuntu. Could you help me out with a little code SauvolaMs example ? For example how could I apply a SauvolaMs on a single page .tiff format and/or on many pages in .tiff stacks ? best regards Albert Berman, MSc Ben-Gurion University of the Negev Phone: +972-8-6477320 (office) Phone: +972-8-6477311 (lab) Fax: +972-8-6477627 P.O.B. 653 Beer Sheva 84105 Israel

2 1

milena_one.cc:14:25: fatal error: tests/data.hh: No such file or directory
by Albert Berman 05 Sep '13

05 Sep '13

Helo epita, I trying to get running the olena on ubuntu in order to test SauvolaMs myself, but by compiling the"3.2 First generic algorithm" from the tutorial I am always getting the same error : milena_one.cc:14:25: fatal error: tests/data.hh: No such file or directory what I am doing wrong ? best regards Albert Berman

2 1

Branching and committing policy change
by Roland Levillain 02 Sep '13

02 Sep '13

Hi everybody, Starting today, we are changing our branching and committing policy. The branch `master', which used to be used to tag the latest release (currently, Olena 2.0) will be used as a stable integration branch. This role was previously played by the branch `next'. This change is motivated by the fairly recent introduction of TeamCity (http://teamcity.lrde.epita.fr) as continuous integration solution at LRDE and by the fact that `master' should provide more up-to-date (yet stable) contents. As this was the case for `next', no patches should be push directly into `master'. (Actually, the Git central repository will forbid any change to `master', except for members of the maintainer team, which is currently reduced to me & myself.) New changes shall be sent through another branch (either an existing one or a new one), preferably on top of `master' to make future integration easier. Any integration effort into `master' shall start with a pull request, either informally, by contacting a (the) maintainer, or better, by creating a pull-request ticket on Trac (use https://trac.lrde.epita.fr/olena/newticket and set the type to `pull-request'). Patches will be reviewed and integrated trough a merge or a rebase action onto `master'. Things will get a little clearer and simpler when GitLab is officially deployed and used at LRDE. The fate of the branch `next' has not been completely decided yet. It will probably be used as an antechamber for `master', for short fix patches only. Other contributions should be recorded into branches with relevant names, so as to make branch management easier. I'll try to put more useful information regarding Git usage in `HACKING' (http://git.lrde.epita.fr/?p=olena.git;a=blob;f=HACKING) ASAP. -- Roland Levillain EPITA Research and Development Laboratory (LRDE) 14-16, rue Voltaire - FR-94276 Le Kremlin-Bicêtre Cedex - France Phone: +33 1 53 14 59 45 - Fax: +33 1 53 14 59 22 - www.lrde.epita.fr

1 0