[Scilab-users] What is the future of Scilab?
Antoine Monmayrant
antoine.monmayrant at laas.fr
Wed Feb 17 14:34:55 CET 2016
Le 02/17/2016 02:24 PM, Philipp Mühlmann a écrit :
> Quote:
>
> ''Concerning the first issue, image processing with scilab, I think a
> basic core image support should be part of scilab.''
>
> Well, agreed to this.
>
>
> However here a different approach:
>
> I wonder if it couldn't be helpful to connect Scilab with a native
> Image processing Programm...like Inkscape.
>
> Why trying to convert Scilab into a image processing tool, while there
> are dedicated programs for it?
>
> Yes I know: It might be better to have everything in one package,
> beeing independent from other Software.
> As I use IPD very much I'm glad to have this available.
>
> I just have to think about a discussion about one complaining about
> Images, trying to manage his images with MS-Word instead of using a
> proper Software.
>
>
> Basic Image processing functions like rotate, shift, etc. might be
> done using the "dos" command and Inkscape as well.
Hi Philipp,
I think we are not talking about the same kind of processing here.
Image processing in the context of Scilab is more opencv-like than
inkscape-like.
I want to import into scilab an image (3*8bits RGB bmp, 3*32bits RGB
bmp, 16bits grayscale TIFF, ...) and map it to the most relevant
variable in scilab (HxWX3 hypermatrix of uint8 for HxW 8bitRGB bmp,
...), in order to do some number crunching on the data contained in the
image.
It's not really cropping, rotating or some sort of "photoshoping" that I
need, it's basically a way to "process" (as in signal processing).
Cheers,
Antoine
>
>
> BR
> Philipp
>
>
>
>
>
>
>
>
>
>
>
>
>
> 2016-02-17 12:21 GMT+01:00 Alasdair McAndrew <amca01 at gmail.com
> <mailto:amca01 at gmail.com>>:
>
> The poor quality of image processing toolboxes is what turned me
> off Scilab to GNU Octave, which has a very good imaging toolbox.
> Or Python, which with its various imaging libraries and
> scientific/numeric libraries is now a serious competitor for
> matrix-oriented software like Matlab/Scilab/Octave. I've written
> two image processing textbooks, and at no time has Scilab ever
> been a serious contender for their software.
>
> On Wed, Feb 17, 2016 at 7:20 PM, <amonmayr at laas.fr
> <mailto:amonmayr at laas.fr>> wrote:
>
> Le 02/17/2016 01:23 AM, AlvaroOsvaldo a écrit :
>> I am worried, in previous versions of scilab, I implemented a
>> medium project for scientific computing with scilab at that
>> time had many memory leaks and had been a nightmare. With the
>> new updates that was fixed and this much better. But now I
>> am implementing another project involving massive image
>> processing. And the three atoms for this, one does not work,
>> one does not load images but the manual says they can do, and
>> other charges but leak the computer's memory and it took me
>> longer skirting the problem than implementing the solution.
>> And worse, because this the system is much slower than it
>> should. This atom memory failure it has been known, however,
>> despite it make impraticable to use the atom for medium and
>> large projects, makes three years that no one gets the bug.
>> forge.scilab.org/index.php/p/IPD/issues/992/
>> <http://forge.scilab.org/index.php/p/IPD/issues/992/> There
>> are high expectations with Scilab 6 and I believe we will
>> have a much better platform. But for what you know, in the
>> meantime we will run out of many atoms, and this greatly
>> reduces the possibility of working in scilab. What is the
>> future of Scilab? For now, the impression I have is that the
>> atoms are abandoned and without the atoms the usefulness of
>> scilab is reduced a lot.
>
> Hi,
>
> I think there are two issues here:
>
> 1) The sad state of image processing toolbox with Scilab. It's
> terrrible. Image processing with Scilab is a nightmare. Every
> other language I use is far better than Scilab in this area
> (even extremely young language like Julia offers a better
> experience. The 3 toolboxes are either broken, unmaintained or
> difficult (or impossible) to install. And the experience is
> even worse under Linux (IPD for example requires a very old
> version of opencv to be manually installed under Linux which
> is a daunting task for most user and which introduces many
> compatibility issues with other softwares relying on opencv).
>
> 2) The lack of information on the life (or lack of life) for a
> given atoms module. I know some information is available on
> the atoms website, but I've always found it difficult to
> determine whether the package was long dead or under active
> development. There are some long dead modules that are still
> listed alongside with actively maintained ones and that show
> excellent rating. It's a bit confusing.
>
> Concerning the first issue, image processing with scilab, I
> think a basic core image support should be part of scilab.
>
> Antoine
>
>
>
>
>> ------------------------------------------------------------------------
>> View this message in context: What is the future of Scilab?
>> <http://mailinglists.scilab.org/What-is-the-future-of-Scilab-tp4033471.html>
>> Sent from the Scilab users - Mailing Lists Archives mailing
>> list archive
>> <http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html>
>> at Nabble.com.
>>
>>
>> _______________________________________________
>> users mailing list
>> users at lists.scilab.org <mailto:users at lists.scilab.org>
>> http://lists.scilab.org/mailman/listinfo/users
>
>
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org <mailto:users at lists.scilab.org>
> http://lists.scilab.org/mailman/listinfo/users
>
>
>
>
> --
> http://www.facebook.com/alasdairmcandrew
> https://twitter.com/amca01 <https://twitter.com/amca01>
> <https://www.linkedin.com/in/alasdair-mcandrew-108178a>
> <https://plus.google.com/+AlasdairMcAndrew/posts>
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org <mailto:users at lists.scilab.org>
> http://lists.scilab.org/mailman/listinfo/users
>
>
>
>
> --
> There we have the salad.
--
+++++++++++++++++++++++++++++++++++++++++++++++++++++++
Antoine Monmayrant LAAS - CNRS
7 avenue du Colonel Roche
BP 54200
31031 TOULOUSE Cedex 4
FRANCE
Tel:+33 5 61 33 64 59
email : antoine.monmayrant at laas.fr
permanent email : antoine.monmayrant at polytechnique.org
+++++++++++++++++++++++++++++++++++++++++++++++++++++++
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20160217/4da81333/attachment.htm>
More information about the users
mailing list