[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: imread (repost)
From: |
John W. Eaton |
Subject: |
Re: imread (repost) |
Date: |
Tue, 05 Aug 2008 23:05:35 -0400 |
On 5-Aug-2008, Thomas L. Scofield wrote:
|
| In Octave:
|
| octave:4> im = imread("magnolia.png");
| octave:5> whos im
| Variables in the current scope:
|
| Attr Name Size Bytes Class
| ==== ==== ==== ===== =====
| im 480x640x4 1228800 uint8
|
| Total is 1228800 elements using 1228800 bytes
|
| octave:6> max(max(im(:,:,4)))
| ans = 247
| octave:7> min(min(im(:,:,4)))
|
|
| octave:8> [im,map,alpha] = imread("magnolia.png");
| octave:9> whos im map alpha
| Variables in the current scope:
|
| Attr Name Size Bytes Class
| ==== ==== ==== ===== =====
| im 480x640x4 1228800 uint8
| map 0x0 0 double
| alpha 0x0 0 double
|
| Total is 1228800 elements using 1228800 bytes
|
|
| In Matlab:
|
| >> im = imread('magnolia.png');
| >> whos im
| Name Size Bytes Class Attributes
|
| im 480x640 307200 uint8
Hmm. that seems like a bug. Or is it documented somewhere? And waht
version of Matlab are you using? Can you send the image to me?
| >> [im, map, alpha] = imread('magnolia.png');
| >> whos im map alpha
| Name Size Bytes Class Attributes
|
| alpha 0x0 0 double
| im 480x640 307200 uint8
| map 250x3 6000 double
|
|
| For this same image, here is the output of the GraphicsMagick shell
| command "gm identify -verbose magnolia.png":
|
| Image: /Users/scofield/images/generic/magnolia.png
| Format: PNG (Portable Network Graphics)
| Geometry: 640x480
| Class: DirectClass
| Type: true color with transparency
| Depth: 8 bits-per-pixel component
| Channel Depths:
| Red: 8 bits
| Green: 8 bits
| Blue: 8 bits
| Opacity: 8 bits
OK. First, it seems that imread needs some changes so that when im
and map are requested, we return the image in that format instead of
just ignoring those outputs.
Given that Matlab doesn't seem to handle alpha in this case, I'm not
sure what is best (whatever we do, there is some chance that we will
have to change later for compatibility reasons).
jwe
- imread (repost), Thomas L. Scofield, 2008/08/04
- Re: imread (repost), Søren Hauberg, 2008/08/05
- Re: imread (repost), Thomas L. Scofield, 2008/08/05
- Re: imread (repost), Søren Hauberg, 2008/08/05
- Re: imread (repost), Søren Hauberg, 2008/08/05
- Re: imread (repost), Thomas L. Scofield, 2008/08/05
- Re: imread (repost), John W. Eaton, 2008/08/05
- Re: imread (repost), Thomas L. Scofield, 2008/08/05
- Re: imread (repost),
John W. Eaton <=
- Re: imread (repost), John W. Eaton, 2008/08/06
- Re: imread (repost), Thomas L. Scofield, 2008/08/06
- Re: imread (repost), John W. Eaton, 2008/08/06
- Re: imread (repost), Bill Denney, 2008/08/05