[Scilab-users] IPCV upscaling of image

[Claus Futtrup]

Hi Philipp

Interesting ... I think this should be part of the help available.
I hope Chin Lun Tan sees this and feel inspired...

Cheers,
Claus

On 21-12-2020 21:54, P M wrote:
> Hi Claus,
>
> this works for me:
>
> imgList  =  list();
> imgList(1)  =  imread(fullpath(getIPCVpath()  +  "/images/superres/input001.png"));
> imgOut   =  imsuperres(imgList);
> figure();
> imshow(imgOut);
> outPath_1  =  'd:\testIn.png';
> outPath_2  =  'd:\testOut.png';
> res  =  imwrite(imgList(1),  outPath_1);
> res  =  imwrite(imgOut,  outPath_2);
> xdel();
> The input file is saved as a 64x64 image.
> The superresolution image is saved as a 256 x 256 image.
>
> Interestingly enough:
> If the output paths are set to "c:\testOut.png", imwrite would not 
> work...unsure why.
>
> Regards, Philipp
>
> Am Mo., 21. Dez. 2020 um 11:37 Uhr schrieb Claus Futtrup 
> <cfuttrup at gmail.com <mailto:cfuttrup at gmail.com>>:
>
>     Hi Philipp
>
>     The file was saved automatically. I don't think you can set what
>     to save, or even the name of the file.
>     I also did a xs2png, but that only saves the current figure. In my
>     case a 610 x 460 pixel image (=the screen figure).
>
>     Cheers,
>     Claus
>
>     On 21-12-2020 11:21, P M wrote:
>>     Dear Claus,
>>
>>     the example did not include saving the image as a file.
>>
>>     Best guesses:
>>     - use imwrite
>>     - use one of the xs2...-functions (e.g.: xs2bmp)
>>
>>     BR
>>     Philipp
>>
>>
>>     Am Mo., 21. Dez. 2020 um 09:20 Uhr schrieb Claus Futtrup
>>     <cfuttrup at gmail.com <mailto:cfuttrup at gmail.com>>:
>>
>>         Hi Philipp
>>
>>         I see, thank you for explaining it to me. I also see that the
>>         upscale is not 2x2 = 4, but it is 4x4 so the upscaled image
>>         is huge. The output file is saved - I believe - in wherever
>>         your Scilab file browser is currently active.
>>
>>         Best regards,
>>         Claus
>>
>>         On 20-12-2020 22:55, P M wrote:
>>>         Dear Claus,
>>>
>>>         you need to feed a list into imsuperres()....not directly
>>>         the image.
>>>         Following should work.
>>>         imgList  =  list();
>>>         imgList(1)  =  imread('path_to_image');
>>>         imgOut   =  imsuperres(imgList);figure(); imshow(imgList(1)); figure();
>>>         imshow(imgOut); BR Philipp
>>>
>>>
>>>         Am So., 20. Dez. 2020 um 19:42 Uhr schrieb Claus Futtrup
>>>         <cfuttrup at gmail.com <mailto:cfuttrup at gmail.com>>:
>>>
>>>             Hi there
>>>
>>>             I have a 4K picture, which I thought I'll try to upscale
>>>             with the
>>>             imsuperres in the IPCV toolbox. The code simply looks
>>>             like this:
>>>
>>>             imagefile =
>>>             'D:\Userdata\Claus\Pictures\Think_Global_4K.png';
>>>             im = imread(imagefile);
>>>             imout  = imsuperres(im);
>>>
>>>             The file is read and I get the following response:
>>>             Unsigned Integer 8
>>>             bits (2160 x 3840). Then Scilab crashes. I'm running
>>>             Scilab 6.1.0 (the
>>>             original release) on Windows 10 - in a MS Windows dialog
>>>             box I get:
>>>
>>>              >Scilab 6.1.0 (Desktop) has stopped working
>>>
>>>              >A problem caused the program to stop working
>>>             correctly. Windows will
>>>             close the program and notify you if a solution is available.
>>>
>>>             I am not sure that I've understood how to use imsuperres
>>>             correctly. I
>>>             see in the _params that rfactor = 4 is the default, so I
>>>             was expecting
>>>             the image to be scaled by a factor 4 (maybe that's 2 x 2
>>>             = I get an 8K
>>>             image = 4320 x 7680 pixels). The way I understand the
>>>             documentation, I
>>>             don't need to run imsuperres_params if I'm satisified
>>>             with the defaults.
>>>
>>>             Kind regards,
>>>
>>>             Claus
>>>
>>>             _______________________________________________
>>>             users mailing list
>>>             users at lists.scilab.org <mailto:users at lists.scilab.org>
>>>             http://lists.scilab.org/mailman/listinfo/users
>>>             <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://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://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://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://lists.scilab.org/mailman/listinfo/users>
>
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org
> http://lists.scilab.org/mailman/listinfo/users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20201222/6b01e045/attachment.htm>