octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #41240] image class is not preserved in cdata


From: Carnë Draug
Subject: [Octave-bug-tracker] [bug #41240] image class is not preserved in cdata field of graphics handle
Date: Wed, 15 Jan 2014 17:43:23 +0000
User-agent: Mozilla/5.0 (X11; Linux i686; rv:17.0) Gecko/20131030 Firefox/17.0 Iceweasel/17.0.10

URL:
  <http://savannah.gnu.org/bugs/?41240>

                 Summary: image class is not preserved in cdata field of
graphics handle
                 Project: GNU Octave
            Submitted by: carandraug
            Submitted on: Wed 15 Jan 2014 17:43:22 GMT
                Category: None
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Other
                  Status: None
             Assigned to: None
         Originator Name: 
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: 3.8.0
        Operating System: Any

    _______________________________________________________

Details:

The data of an image being plotted is kept in the field cdata, but the class
has changed. This is important because the class of an image changes how the
data is interpreted.

For example:


## image is uint8
img = repmat (uint8 (0:250), [200 1]);
h = imshow (img);
## image comes back as double
nimg = get (h, 'cdata');
class (nimg)
## note how changing class has a drastic effect on its representation
figure ();
imshow (nimg);


Note how the the same function, imshow, displays the image so differently. As
uint8, it's a gradient, as double, it's all white (since images of class
double should be on range [0 1]).

The effect this has on other functions is that it's not possible to retrieve
the image from a figure (required for things such as impixel and imcrop which
return parts from the current or a specified figure).

I looked at the other fields available from get() but didn't notice anything
that would point to the original class.

Also, this works in MatLab so it stacks as a matlab compatibility bug.




    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?41240>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]