From fredrik at pythonware.com  Sun Oct  2 17:37:36 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 2 Oct 2005 17:37:36 +0200
Subject: [Image-SIG] Usage of getpixel
References: <000a01c5c2b0$7fa37aa0$057d7053@dell>
Message-ID: <dhous2$i70$1@sea.gmane.org>

"alainlioret" <alainlioret at wanadoo.fr> wrote:

> I try to use getpixel(xy) ...
>
> But I can't ..
>
> for example, I try :
>
> pix = im.getpixel(0,0)
>
> But I have an error. What's wrong ?
>
> Can you give me an example ?

getpixel expects a coordinate tuple, not two separate arguments.  to
get the pixel in the upper left corner, use:

    pix = im.getpixel((0, 0))

</F>




From x12345jp at yahoo.com.hk  Tue Oct  4 16:30:57 2005
From: x12345jp at yahoo.com.hk (john taylor)
Date: Tue, 4 Oct 2005 22:30:57 +0800 (CST)
Subject: [Image-SIG] How to deal with BGRA color model in PIL?
Message-ID: <20051004143057.57075.qmail@web50415.mail.yahoo.com>

hi all,

i have a string of pixel values, which uses the BGRA
color model, is there an easy way (hopefully the most
efficient way) to convert it into the RGBA model, so
that the image can be displayed correctly?

thanks a lot!

john

From gandalf at designaproduct.biz  Tue Oct  4 16:46:39 2005
From: gandalf at designaproduct.biz (Laszlo Zsolt Nagy)
Date: Tue, 04 Oct 2005 16:46:39 +0200
Subject: [Image-SIG] How to deal with BGRA color model in PIL?
In-Reply-To: <20051004143057.57075.qmail@web50415.mail.yahoo.com>
References: <20051004143057.57075.qmail@web50415.mail.yahoo.com>
Message-ID: <434295CF.2010902@designaproduct.biz>

john taylor wrote:

>hi all,
>
>i have a string of pixel values, which uses the BGRA
>color model, is there an easy way (hopefully the most
>efficient way) to convert it into the RGBA model, so
>that the image can be displayed correctly?
>  
>
b,g,r,a = original_image.split()
new_image= Image.merge("RGBA", (r,g,b,a))

BTW, there is an almost identical example in the documentation, section 
"Cutting, Pasting and Merging Images / Example: Splitting and merging 
bands" - please read the docs.

   Les

From x12345jp at yahoo.com.hk  Wed Oct  5 16:41:08 2005
From: x12345jp at yahoo.com.hk (john taylor)
Date: Wed, 5 Oct 2005 22:41:08 +0800 (CST)
Subject: [Image-SIG] error in ImageChops.logical_xor ???
Message-ID: <20051005144108.61506.qmail@web50407.mail.yahoo.com>

hi all, 

i want to compare two images, they are of the same
mode ('RGB'), same format ('BMP'), same size, etc. 

i am using ImageChops.logical_xor(img1, img2), but it
gives me this error:

Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File
"C:\Python24\Lib\site-packages\PIL\ImageChops.py",
line 262, in logical_xor
    return image1._new(image1.im.chop_xor(image2.im))
ValueError: image has wrong mode

i've also tried calling img1.im.chop_xor(img2.im), and
not surprisingly i get:

Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
ValueError: image has wrong mode

anybody can help on this? and which is the fastest way
to compare two images, if i only want to know if they
are identical or not?

thanks a lot!
john

From fredrik at pythonware.com  Wed Oct  5 21:23:05 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 5 Oct 2005 21:23:05 +0200
Subject: [Image-SIG] error in ImageChops.logical_xor ???
References: <20051005144108.61506.qmail@web50407.mail.yahoo.com>
Message-ID: <di196l$kp9$1@sea.gmane.org>

"john taylor" wrote:

> i want to compare two images, they are of the same
> mode ('RGB'), same format ('BMP'), same size, etc.

here's a reasonably fast way to compare two images:

def compare(im1, im2):
    import ImageChops
    if im1.mode != im2.mode or im1.size != im2.size:
        return 0
    try:
        return ImageChops.difference(im1, im2).getbbox() is None
    except ValueError:
        return im1.tostring() == im2.tostring()

I think the logical operators (which are undocumented) only work on
mode "1" images.

</F> 




From fredrik at pythonware.com  Thu Oct  6 13:55:58 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 6 Oct 2005 13:55:58 +0200
Subject: [Image-SIG] How to deal with BGRA color model in PIL?
References: <20051004143057.57075.qmail@web50415.mail.yahoo.com>
Message-ID: <di33ce$t4m$1@sea.gmane.org>

"john taylor" wrote:

> i have a string of pixel values, which uses the BGRA
> color model, is there an easy way (hopefully the most
> efficient way) to convert it into the RGBA model, so
> that the image can be displayed correctly?

if you have data in a string, you can use Image.fromstring to load it into a
memory buffer, converting it on the way:

    import Image

    pixel = "\x01\x02\x03\x04"

    im = Image.fromstring("RGBA", (1, 1), pixel, "raw", "BGRA")

    print im.getpixel((0, 0)) # prints (3, 2, 1, 4)

if you have the data in an image memory in the wrong order, you can either
use split/merge (see Laszlo's reply), or roundtrip via a string buffer:

    im = ...

    new_im = Image.fromstring("RGBA", im.size, im.tostring(), "raw", "BGRA")

</F> 




From b.vinegar at utoronto.ca  Thu Oct  6 22:01:41 2005
From: b.vinegar at utoronto.ca (Ben)
Date: Thu, 06 Oct 2005 16:01:41 -0400
Subject: [Image-SIG] PIL: Flood Color Replace?
Message-ID: <434582A5.5000105@utoronto.ca>

Hey all,

I was wondering if PIL had an efficient function that would let me 
replace all pixels of a specific color with another? I could do it 
myself in Python, but this is for a performance-critical application.

Thanks,
Ben

From fredrik at pythonware.com  Thu Oct  6 22:42:45 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 6 Oct 2005 22:42:45 +0200
Subject: [Image-SIG] Flood Color Replace?
References: <434582A5.5000105@utoronto.ca>
Message-ID: <di4287$clu$1@sea.gmane.org>

"Ben" wrote:

> I was wondering if PIL had an efficient function that would let me
> replace all pixels of a specific color with another? I could do it
> myself in Python, but this is for a performance-critical application.

all pixels of a specific color in the entire image (=point operation), or
in a given region that already has another color (=flood fill).

</F>




From eachand at gmail.com  Tue Oct  4 18:37:43 2005
From: eachand at gmail.com (Aman)
Date: Tue, 4 Oct 2005 09:37:43 -0700
Subject: [Image-SIG] Multi Page Batch Process
Message-ID: <5dc4aea70510040937g6aa98b42v7317491983ad3016@mail.gmail.com>

Hi

Trying to come up with a routine for joing tiff files:

Input : single page G4 tiff files names
Out put : Multipage tiff files


Does anyone have something along these lines as i have hardly donw
development in python.


Thanks in advance
Aman






--
"..I believe that imagination is stronger than knowledge--that myth is more
potent than history. I believe that dreams are more powerful than
facts--That hope always triumphs over experience--That laughter is the only
cure for grief. And I believe that love is stronger than
death..."-------------Robert Fulghum
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051004/390d75ba/attachment.htm

From jhu at metrigenix.com  Fri Oct  7 23:52:51 2005
From: jhu at metrigenix.com (James Hu)
Date: Fri, 7 Oct 2005 17:52:51 -0400
Subject: [Image-SIG] can't show png file with mode "I"
Message-ID: <90DEA7ED93F71A4580CEDF76A02682EA03ACBE@torexch.metrigenix.com>

Hi,

 

I captured image from digital camera and saved it to png file with: 

 

Image.fromstring("I", datasize, origFrame.pBuffer, 'raw', 'I;16')

 

But when I open it  with PIL.Image

 

im=Image.open("*.png")

Im.show()

 

Got white image, 

What might be the problem?

This png file can be open with IrfanView, or using wx.Image.

 

Thanks in advance.

 

 

James

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051007/7536f892/attachment.html

From fredrik at pythonware.com  Sat Oct  8 21:19:02 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 8 Oct 2005 21:19:02 +0200
Subject: [Image-SIG] can't show png file with mode "I"
References: <90DEA7ED93F71A4580CEDF76A02682EA03ACBE@torexch.metrigenix.com>
Message-ID: <di9634$n7n$1@sea.gmane.org>

(stuck in the spam filter, due to excessive use of HTML.  please
post plain text messages to this list if at all possible).

James Hu wrote:

> I captured image from digital camera and saved it to png file with:
>
> Image.fromstring("I", datasize, origFrame.pBuffer, 'raw', 'I;16')
>
> But when I open it  with PIL.Image
>
> im=Image.open("*.png")
> Im.show()
>
> Got white image,
> What might be the problem?

I've answered this on comp.lang.python:

    http://article.gmane.org/gmane.comp.python.general/424715

but here's the answer again, for the archives:

    maybe all pixels in the PNG file are larger than 255 ?

    show doesn't support 16-bit images, so it clamps the values down to
    an 8-bit range.  adding a

        print im.getextrema()

    will tell you what values you have in the image (it returns min and max).

    to scale things down before you display it, you can use something like:

        im = im.point(lambda x: x*(1.0/256))

    or

        lo, hi = im.getextrema()
        if lo == hi:
            im = im.point(lambda x: 128) # or something
        else:
            scale = 255.0 / (hi - lo)
            offset = -lo * scale + 0.5
            im = im.point(lambda x: x*scale + offset)

hope this helps!

</F>




From michele.petrazzo at unipex.it  Mon Oct 10 09:39:12 2005
From: michele.petrazzo at unipex.it (Michele Petrazzo)
Date: Mon, 10 Oct 2005 09:39:12 +0200
Subject: [Image-SIG] Multi Page Batch Process
In-Reply-To: <5dc4aea70510040937g6aa98b42v7317491983ad3016@mail.gmail.com>
References: <5dc4aea70510040937g6aa98b42v7317491983ad3016@mail.gmail.com>
Message-ID: <434A1AA0.6030908@unipex.it>

Aman wrote:
> Hi
> 
> Trying to come up with a routine for joing tiff files:
> 
> Input : single page G4 tiff files names Out put : Multipage tiff 
> files
> 
> 
> Does anyone have something along these lines as i have hardly donw 
> development in python.
> 

I don't know if you know freeimagepy, but I develop it because I didn't
find any other project that work well with tiff G3/G4 and multipage files.
Try it, if you want of course :), and let me know if you have problems.

freeimagepy.sf.net

> 
> Thanks in advance Aman
> 

Michele

From daviddench at tiscali.co.uk  Mon Oct 10 14:07:54 2005
From: daviddench at tiscali.co.uk (David Dench)
Date: Mon, 10 Oct 2005 13:07:54 +0100
Subject: [Image-SIG] Image Stacking and AVI processing
Message-ID: <434A599A.10207@tiscali.co.uk>

Dear All,
I am returning to the fold after a long absence and need a quick start.
I want to split an AVI into a set of cropped AVI's  eg split the AVI 
from 640x480 sized frames
into 4 quarter sized AVIs  of same length but smaller frame size.
I then want to stack/average each AVI ( usually after aligning on a 
feature ) and then recombine
the stacked images back into a single full size 640x480 image.
Some of you may recognize similarities to Registax functionality with 
good reason.
I have done a quick Python trawl and found PyMedia and NumArray which 
might help
in what I want to do ( I know PIL of old and expect to use it ).
Have I missed any existing software in my ignorance or are there 
alternative packages that
I should be investigating ?
    Very best wishes to all,
                David Dench
PS As an alternative, I could glue together some existing Windows GUI 
intensive Apps. Is there a python
Expect like package suitable for the task ?

From gwidion at mpc.com.br  Mon Oct 10 15:08:27 2005
From: gwidion at mpc.com.br (Joao S. O. Bueno Calligaris)
Date: Mon, 10 Oct 2005 10:08:27 -0300
Subject: [Image-SIG] Image Stacking and AVI processing
In-Reply-To: <434A599A.10207@tiscali.co.uk>
References: <434A599A.10207@tiscali.co.uk>
Message-ID: <200510101008.27571.gwidion@mpc.com.br>

On Monday 10 October 2005 09:07, David Dench wrote:
> Dear All,
> I am returning to the fold after a long absence and need a quick
> start. I want to split an AVI into a set of cropped AVI's  eg split
> the AVI from 640x480 sized frames
> into 4 quarter sized AVIs  of same length but smaller frame size.
> I then want to stack/average each AVI ( usually after aligning on a
> feature ) and then recombine
> the stacked images back into a single full size 640x480 image.
> Some of you may recognize similarities to Registax functionality
> with good reason.
> I have done a quick Python trawl and found PyMedia and NumArray
> which might help
> in what I want to do ( I know PIL of old and expect to use it ).
> Have I missed any existing software in my ignorance or are there
> alternative packages that
> I should be investigating ?
>     Very best wishes to all,
>                 David Dench
> PS As an alternative, I could glue together some existing Windows
> GUI intensive Apps. Is there a python
> Expect like package suitable for the task ?

This sounds  "doable" with the Gimp Animation Package (GIMP-GAP)  
extension for the GIMP. It won't be trivial, though.
If you want to try do do it interactively instead of programatically 
(you can code python scripts for intermediate steps in the proccess 
if you need to), go for it.

> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig

From fredrik at pythonware.com  Sat Oct 15 18:18:51 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 15 Oct 2005 18:18:51 +0200
Subject: [Image-SIG] AGG-based drawing for PIL
References: <d966j5$rj4$1@sea.gmane.org> <dh64pa$o15$1@sea.gmane.org>
Message-ID: <dira5c$tl2$1@sea.gmane.org>

> the "aggdraw" module is a prerelease of a new drawing library for PIL,
> based on Maxim Shemanarev's antigrain (AGG) graphics library.  This
> module implements a WCK-style drawing interface (see below), and
> provides anti-aliasing and alpha compositing.

1.1 final is now available from:

    http://effbot.org/downloads/#aggdraw

documentation:

    http://effbot.org/zone/draw-agg.htm (overview)
    http://effbot.org/zone/pythondoc-aggdraw.htm (api reference)

for a list of changes in this release, see:

    http://effbot.org/downloads/index.cgi/aggdraw-1.1-20051010.zip/CHANGES

enjoy /F




From dima at shiftingplanes.org  Sun Oct 16 01:40:16 2005
From: dima at shiftingplanes.org (Dmitry (Dima) Strakovsky)
Date: Sat, 15 Oct 2005 18:40:16 -0500
Subject: [Image-SIG] vector graphics Q
Message-ID: <43519360.4040203@shiftingplanes.org>

hi all,

My question is perhaps related to Fredrik earlier post. I read API doc 
(understood about 30%)

I am basically looking for a way to manupulate vector graphics in 
Python. So far I have been reading up on SVG and looking at the output 
from Illustrator and Inkscape (Adobe seems to throw a whole bunch of 
proprietary stuff into the files, surprise?not) The images for this 
particular project are going to be generated in Illustrator (the program 
I am most familiar with) and probably cleaned up with another utility 
(possibly python script), after which I was going to do a fair amount of 
manipulation (hopefully python script) and the final output goes to the 
printer (rastered.)

Fredrick does your lib work with SVG?
Is there a library out there that is particularly nice (has 
resize,rotate,copy,paste type functions)for working with SVGs?
Am I married to SVG for absolutely no reason?


As I stated in an earlier post I am new to both Python and PIL (and a 
fairly crapy (read not a professional) programmer on top of that :) so 
links to source code snippets are much appreciated :)  <-- and  NO there 
will not be any write-this-whole-program-for-me type follow up questions!

                                                                thnx in 
advance,
                                                                         
                           dima

From p.f.moore at gmail.com  Sun Oct 16 14:54:19 2005
From: p.f.moore at gmail.com (Paul Moore)
Date: Sun, 16 Oct 2005 13:54:19 +0100
Subject: [Image-SIG] vector graphics Q
In-Reply-To: <43519360.4040203@shiftingplanes.org>
References: <43519360.4040203@shiftingplanes.org>
Message-ID: <79990c6b0510160554u6d93597fi24ec97c736161996@mail.gmail.com>

On 10/16/05, Dmitry (Dima) Strakovsky <dima at shiftingplanes.org> wrote:
> Fredrick does your lib work with SVG?

I believe that PIL is for bitmap images only, and as such, SVG isn't
supported. If you look in the appendix "Image File Formats" of the PIL
handbook (http://www.pythonware.com/library/pil/handbook/index.htm)
you'll see that SVG isn't documented as being supported.

> Is there a library out there that is particularly nice (has
> resize,rotate,copy,paste type functions)for working with SVGs?

I also have some interest in vector graphics processing. (Although
it's fairly casual so far).

One specific thing I would like to be able to do is to take a bitmap
format of a line image (a scanned copy of a line drawing, to be
specific) and convert it to vector format. I realise that this isn't a
trivial task (edge detection, and so on) but are there any tools or
libraries that will do this? I haven't found any support in PIL for
this, but maybe I've missed something...

Thanks,
Paul.

From dima at shiftingplanes.org  Sun Oct 16 15:39:48 2005
From: dima at shiftingplanes.org (Dmitry (Dima) Strakovsky)
Date: Sun, 16 Oct 2005 08:39:48 -0500
Subject: [Image-SIG] vector graphics Q
In-Reply-To: <79990c6b0510160554u6d93597fi24ec97c736161996@mail.gmail.com>
References: <43519360.4040203@shiftingplanes.org>
	<79990c6b0510160554u6d93597fi24ec97c736161996@mail.gmail.com>
Message-ID: <43525824.50707@shiftingplanes.org>

Hi Paul,

I haven't done any testing of bitmap to SVG myself, so this is not a 
voice of experience :)
hope this helps:

http://www.scale-a-vector.de/svg-test4-e.htm
http://autotrace.sourceforge.net/

Would love to hear if people on the list have used any of the above tools.

Actually handbook is what got me using PIL (thank you Fredrik,Matthew)  
it is one of the BEST features of the library. I tried using 
ImageMagick  before and encountered severe lack of well written 
documentation (and lots of crazy uses for pointers and macros.)  ouch!  
However the handbook was written almost 3 years ago and while the core 
of the library has not changed (i think) some new features might have 
been added (as demonstrated by Fredrik's post yesterday.) I know I am 
being a noob, but I figure I gotta ask :)

I found the following lib
http://www2.sfk.nl/svg

... and once again hit a documentation wall (not as severe in python 
case.) Has anyone used this lib? It looks like development on it has 
stopped sometime last year. Is there something comparable brewing 
somewhere else?

Thank you very much for your reply Paul and good luck with bitmap to 
SVG. If I come accross any more info I'll post it to the list. 

                                        dima

Paul Moore wrote:

>On 10/16/05, Dmitry (Dima) Strakovsky <dima at shiftingplanes.org> wrote:
>  
>
>>Fredrick does your lib work with SVG?
>>    
>>
>
>I believe that PIL is for bitmap images only, and as such, SVG isn't
>supported. If you look in the appendix "Image File Formats" of the PIL
>handbook (http://www.pythonware.com/library/pil/handbook/index.htm)
>you'll see that SVG isn't documented as being supported.
>
>  
>
>>Is there a library out there that is particularly nice (has
>>resize,rotate,copy,paste type functions)for working with SVGs?
>>    
>>
>
>I also have some interest in vector graphics processing. (Although
>it's fairly casual so far).
>
>One specific thing I would like to be able to do is to take a bitmap
>format of a line image (a scanned copy of a line drawing, to be
>specific) and convert it to vector format. I realise that this isn't a
>trivial task (edge detection, and so on) but are there any tools or
>libraries that will do this? I haven't found any support in PIL for
>this, but maybe I've missed something...
>
>Thanks,
>Paul.
>
>
>  
>


From fredrik at pythonware.com  Sun Oct 16 18:03:00 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 16 Oct 2005 18:03:00 +0200
Subject: [Image-SIG] vector graphics Q
References: <43519360.4040203@shiftingplanes.org>
Message-ID: <dittjm$l1a$1@sea.gmane.org>

Dmitry (Dima) Strakovsky wrote:

> My question is perhaps related to Fredrik earlier post. I read API doc
> (understood about 30%)
>
> I am basically looking for a way to manupulate vector graphics in
> Python. So far I have been reading up on SVG and looking at the output
> from Illustrator and Inkscape (Adobe seems to throw a whole bunch of
> proprietary stuff into the files, surprise?not) The images for this
> particular project are going to be generated in Illustrator (the program
> I am most familiar with) and probably cleaned up with another utility
> (possibly python script), after which I was going to do a fair amount of
> manipulation (hopefully python script) and the final output goes to the
> printer (rastered.)
>
> Fredrick does your lib work with SVG?

aggdraw and PIL only provide a small part of the full "graphics application"
stack.  a full "stack" (e.g. Illustrator) might look like this:

    1. a user interface
    2. a data model that represents the drawing
    3. an I/O library that can be used to store the drawing on disk
    4. a rendering system

in your case, you want to replace the user interface with scripts, so you
need a "stack" that looks like this:

    1. a script engine interface
    2. a data model that represents the drawing
    3. an I/O library that can be used to store the drawing on disk
    4. a rendering system

the aggdraw library provides a (somewhat limited) implementation of 4,
and some support for 3 (it can parse SVG path descriptors), but nothing
more than that.

if aggdraw/PIL/Python is a realistic approach depends on what kind of
drawings you will work on, and what you need to do it with them.  I've
been using the following toolchain for a couple of projects:

    1. draw in illustrator
    2. if necessary, convert to outlines
    3. save as SVG
    4. load into element structure using ElementTree:
            http://effbot.org/zone/element-index.htm
    5. manipulate element structure
    6. render relevant parts using aggdraw
    7. save using PIL

this is quite useful for simple drawings, such as icons and other symbols,
but might need a lot of work if you want to support more illustrator/SVG
features (and/or get rid of the second step).

</F>




From p.f.moore at gmail.com  Sun Oct 16 18:44:33 2005
From: p.f.moore at gmail.com (Paul Moore)
Date: Sun, 16 Oct 2005 17:44:33 +0100
Subject: [Image-SIG] vector graphics Q
In-Reply-To: <43525824.50707@shiftingplanes.org>
References: <43519360.4040203@shiftingplanes.org>
	<79990c6b0510160554u6d93597fi24ec97c736161996@mail.gmail.com>
	<43525824.50707@shiftingplanes.org>
Message-ID: <79990c6b0510160944m264944b0ja0707394c338a1f6@mail.gmail.com>

On 10/16/05, Dmitry (Dima) Strakovsky <dima at shiftingplanes.org> wrote:
> I haven't done any testing of bitmap to SVG myself, so this is not a
> voice of experience :)
> hope this helps:
>
> http://www.scale-a-vector.de/svg-test4-e.htm
> http://autotrace.sourceforge.net/

Thanks for the linke - I'll have a look...

Paul.

From Chris.Barker at noaa.gov  Mon Oct 17 07:08:50 2005
From: Chris.Barker at noaa.gov (Chris Barker)
Date: Sun, 16 Oct 2005 22:08:50 -0700
Subject: [Image-SIG] vector graphics Q
In-Reply-To: <43525824.50707@shiftingplanes.org>
References: <43519360.4040203@shiftingplanes.org>	<79990c6b0510160554u6d93597fi24ec97c736161996@mail.gmail.com>
	<43525824.50707@shiftingplanes.org>
Message-ID: <435331E2.9040207@noaa.gov>

You might want to also take a look at wxArt2d. Sadly, however, it not 
yet been wrapped for use with Python.

http://wxart2d.sourceforge.net/

-Chris


-- 
Christopher Barker, Ph.D.
Oceanographer
                                      		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov

From dima at shiftingplanes.org  Wed Oct 19 04:15:55 2005
From: dima at shiftingplanes.org (Dmitry (Dima) Strakovsky)
Date: Tue, 18 Oct 2005 21:15:55 -0500
Subject: [Image-SIG] vector graphics Q
In-Reply-To: <dittjm$l1a$1@sea.gmane.org>
References: <43519360.4040203@shiftingplanes.org> <dittjm$l1a$1@sea.gmane.org>
Message-ID: <4355AC5B.6070807@shiftingplanes.org>

Thanx Fredrik,

I'm slowly making my way through ElementTree code. It's definately an 
intersting direction and fun code to play with :)

Must admit that I am still struggling with OO code, but it is very 
helthy for me to get my head out of assembly mode every once in a while.

          thank you again,

                                                                   dima

-- 
<name>Dmitry (Dima) Strakovsky</name>
<email>dima at shiftingplanes.org</email>
<website>www.shiftingplanes.org</website>



Fredrik Lundh wrote:

>Dmitry (Dima) Strakovsky wrote:
>
>  
>
>>My question is perhaps related to Fredrik earlier post. I read API doc
>>(understood about 30%)
>>
>>I am basically looking for a way to manupulate vector graphics in
>>Python. So far I have been reading up on SVG and looking at the output
>>from Illustrator and Inkscape (Adobe seems to throw a whole bunch of
>>proprietary stuff into the files, surprise?not) The images for this
>>particular project are going to be generated in Illustrator (the program
>>I am most familiar with) and probably cleaned up with another utility
>>(possibly python script), after which I was going to do a fair amount of
>>manipulation (hopefully python script) and the final output goes to the
>>printer (rastered.)
>>
>>Fredrick does your lib work with SVG?
>>    
>>
>
>aggdraw and PIL only provide a small part of the full "graphics application"
>stack.  a full "stack" (e.g. Illustrator) might look like this:
>
>    1. a user interface
>    2. a data model that represents the drawing
>    3. an I/O library that can be used to store the drawing on disk
>    4. a rendering system
>
>in your case, you want to replace the user interface with scripts, so you
>need a "stack" that looks like this:
>
>    1. a script engine interface
>    2. a data model that represents the drawing
>    3. an I/O library that can be used to store the drawing on disk
>    4. a rendering system
>
>the aggdraw library provides a (somewhat limited) implementation of 4,
>and some support for 3 (it can parse SVG path descriptors), but nothing
>more than that.
>
>if aggdraw/PIL/Python is a realistic approach depends on what kind of
>drawings you will work on, and what you need to do it with them.  I've
>been using the following toolchain for a couple of projects:
>
>    1. draw in illustrator
>    2. if necessary, convert to outlines
>    3. save as SVG
>    4. load into element structure using ElementTree:
>            http://effbot.org/zone/element-index.htm
>    5. manipulate element structure
>    6. render relevant parts using aggdraw
>    7. save using PIL
>
>this is quite useful for simple drawings, such as icons and other symbols,
>but might need a lot of work if you want to support more illustrator/SVG
>features (and/or get rid of the second step).
>
></F>
>
>
>
>_______________________________________________
>Image-SIG maillist  -  Image-SIG at python.org
>http://mail.python.org/mailman/listinfo/image-sig
>
>
>  
>


From dima at shiftingplanes.org  Thu Oct 20 06:02:45 2005
From: dima at shiftingplanes.org (Dmitry (Dima) Strakovsky)
Date: Wed, 19 Oct 2005 23:02:45 -0500
Subject: [Image-SIG] possible PIL bug?
Message-ID: <435716E5.2090106@shiftingplanes.org>

Here is a possible PIL bug. I appologise in advance if I am doing 
something silly python-wise.

When using resize method and using BILINEAR, BICUBIC or NEAREST arguments

For example: ImageName.resize((200,300),ANTIALIAS)

interpreter tells me that the global name "ANTIALIAS" is not defined

the globals are definately defined in image.py Shouldn't they be 
imported as a part of
"import Image" routine?

Anyways, entering numeric values:  ImageName.resize((200,300),1)
obviously solves the problem, but makes things a bit harder to read down 
the line

thank you,
                                     dima



-- 
<name>Dmitry (Dima) Strakovsky</name>
<email>dima at shiftingplanes.org</email>
<website>www.shiftingplanes.org</website>


From chris at cogdon.org  Thu Oct 20 06:09:02 2005
From: chris at cogdon.org (Chris Cogdon)
Date: Wed, 19 Oct 2005 21:09:02 -0700
Subject: [Image-SIG] possible PIL bug?
In-Reply-To: <435716E5.2090106@shiftingplanes.org>
References: <435716E5.2090106@shiftingplanes.org>
Message-ID: <EF19A8E5-F4A0-4119-96CD-4E3CB817A2A7@cogdon.org>


On Oct 19, 2005, at 9:02 PM, Dmitry (Dima) Strakovsky wrote:

> Here is a possible PIL bug. I appologise in advance if I am doing
> something silly python-wise.
>
> When using resize method and using BILINEAR, BICUBIC or NEAREST  
> arguments
>
> For example: ImageName.resize((200,300),ANTIALIAS)
>
> interpreter tells me that the global name "ANTIALIAS" is not defined
>
> the globals are definately defined in image.py Shouldn't they be
> imported as a part of
> "import Image" routine?
>
> Anyways, entering numeric values:  ImageName.resize((200,300),1)
> obviously solves the problem, but makes things a bit harder to read  
> down
> the line

Try Image.ANTIALIAS


Python won't 'flood' your namespace with variables unless you  
explicitly ask it to.

-- 
    ("`-/")_.-'"``-._        Chris Cogdon <chris at cogdon.org>
     . . `; -._    )-;-,_`)
    (v_,)'  _  )`-.\  ``-'
   _.- _..-_/ / ((.'
((,.-'   ((,/   fL


From dima at shiftingplanes.org  Fri Oct 21 04:44:45 2005
From: dima at shiftingplanes.org (Dmitry (Dima) Strakovsky)
Date: Thu, 20 Oct 2005 21:44:45 -0500
Subject: [Image-SIG] possible PIL bug?
In-Reply-To: <EF19A8E5-F4A0-4119-96CD-4E3CB817A2A7@cogdon.org>
References: <435716E5.2090106@shiftingplanes.org>
	<EF19A8E5-F4A0-4119-96CD-4E3CB817A2A7@cogdon.org>
Message-ID: <4358561D.2080907@shiftingplanes.org>

Thank for your reply Chris,

Chris Cogdon wrote:

>
> On Oct 19, 2005, at 9:02 PM, Dmitry (Dima) Strakovsky wrote:
>
>> Here is a possible PIL bug. I appologise in advance if I am doing
>> something silly python-wise.
>>
>> When using resize method and using BILINEAR, BICUBIC or NEAREST  
>> arguments
>>
>> For example: ImageName.resize((200,300),ANTIALIAS)
>>
>> interpreter tells me that the global name "ANTIALIAS" is not defined
>>
>> the globals are definately defined in image.py Shouldn't they be
>> imported as a part of
>> "import Image" routine?
>>
>> Anyways, entering numeric values:  ImageName.resize((200,300),1)
>> obviously solves the problem, but makes things a bit harder to read  
>> down
>> the line
>
>
> Try Image.ANTIALIAS
>
>
> Python won't 'flood' your namespace with variables unless you  
> explicitly ask it to.
>
That actually seems like a great language feature. Thank you for 
explaining it.

                                     d



-- 
<name>Dmitry (Dima) Strakovsky</name>
<email>dima at shiftingplanes.org</email>
<website>www.shiftingplanes.org</website>


From fredrik at pythonware.com  Sat Oct 22 15:39:35 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 22 Oct 2005 15:39:35 +0200
Subject: [Image-SIG] AGG-based drawing for PIL
References: <d966j5$rj4$1@sea.gmane.org> <dh64pa$o15$1@sea.gmane.org>
	<dira5c$tl2$1@sea.gmane.org>
Message-ID: <djdfeq$mg0$1@sea.gmane.org>

continuing the tradition of announcing new releases by replying
to myself:

> the "aggdraw" module is a prerelease of a new drawing library for PIL,
> based on Maxim Shemanarev's antigrain (AGG) graphics library.  This
> module implements a WCK-style drawing interface (see below), and
> provides anti-aliasing and alpha compositing.

A very early 1.2 alpha is now available.

You can now create "drawable Windows DIB surfaces", and blit them
to screen under any Windows-based toolkit that lets you access the
HWND for a toolkit window.

    draw = aggdraw.Dib("RGB", (width, height))
    ... draw as usual ...
    draw.expose(hwnd) # blit to screen

Here's a Tkinter example:

    size = width, height

    draw = aggdraw.Dib("RGB", size)
    ... draw as usual ...

    # display the image
    frame = Frame(width=size[0], height=size[1], bg="")
    frame.bind("<Expose>", lambda e: draw.expose(e.widget.winfo_id()))
    frame.pack()

A source kit can be found here:

    http://effbot.org/downloads/#aggdraw

To display aggdraw images under Tkinter on other platforms, you can use
PIL's ImageTk.PhotoImage object, or the WCK toolkit.  See

    http://effbot.org/zone/wck-aggview.htm

for an example on how to use the latter.

</F>




From William.T.Bridgman.1 at gsfc.nasa.gov  Mon Oct 24 17:30:50 2005
From: William.T.Bridgman.1 at gsfc.nasa.gov (W.T. Bridgman)
Date: Mon, 24 Oct 2005 11:30:50 -0400
Subject: [Image-SIG] Undocumented floating-point TIF support in PIL?
Message-ID: <C04529B1-0918-4450-9B87-B6FDBFCCA2D1@gsfc.nasa.gov>

The PIL documentation claims that TIFF files are supported for "1",  
"L", "RGB", or "CMYK" data.

I was examining the Tiff plug-in and did not find anything that  
suggested "F" was unsupported so I tried writing and reading a simple  
floating-point TIFF.  It worked.

Is this just a missing entry in the docs or does anyone know of  
potential gotchas should I start using this new found capability?

Thanks,
Tom
--
Dr. William T."Tom" Bridgman               Scientific Visualization  
Studio
Global Science & Technology, Inc.          NASA/Goddard Space Flight  
Center
Email: William.T.Bridgman.1 at gsfc.nasa.gov  Code 610.3
Phone: 301-286-1346                        Greenbelt, MD 20771
FAX:   301-286-1634                        http://svs.gsfc.nasa.gov/


From janssen at parc.com  Tue Oct 25 02:39:07 2005
From: janssen at parc.com (Bill Janssen)
Date: Mon, 24 Oct 2005 17:39:07 PDT
Subject: [Image-SIG] Image.convert() doesn't dither?
Message-ID: <05Oct24.173910pdt."58617"@synergy1.parc.xerox.com>

The PIL manual says that calling image.convert('1') on a greyscale
image will threshold the image (fairly starkly: "all non-zero values
are set to 255 (white).")  However, when I try this on a test page of
color blocks, I seem to get Floyd-Steinberg dithering instead.

The code sequence I use is:


	import Image, ImageOps

	im = Image.open(infilename)
	if im.mode == 'RGB':
	    im = ImageOps.grayscale(im)
	if im.mode == 'L':
	    im = ImageOps.autocontrast(im)
	    im = im.convert('1')
	im.save(outfilename, 'TIFF')

Is the documentation out of date?

If I want thresholding, is there a now a way to do it without using
"point"?  Using quantize(), perhaps?

Is there a difference between Image.convert('L') and
ImageOps.grayscale()?

Bill

From bbaxter at wadsworth.org  Tue Oct 25 02:47:28 2005
From: bbaxter at wadsworth.org (William Baxter)
Date: Mon, 24 Oct 2005 20:47:28 -0400
Subject: [Image-SIG] ImageTk.py not installed
Message-ID: <435D80A0.9EC24E0B@wadsworth.org>

How come ImageTk.py is not put in the site-packages/PIL directory with
all the other modules? It's there in the
src/Imaging-1.1.5/build/lib.linux-etc directory when you build Imaging.
But it doesn't appear in site-packages/PIL during the install. The
answer is simply to copy it to the place where it's needed, but this has
been a 'gotcha' everytime I've installed Imaging.

Bill Baxter

From fredrik at pythonware.com  Tue Oct 25 11:04:32 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 25 Oct 2005 11:04:32 +0200
Subject: [Image-SIG] Image.convert() doesn't dither?
References: <05Oct24.173910pdt."58617"@synergy1.parc.xerox.com>
Message-ID: <djksf6$6mm$1@sea.gmane.org>

Bill Janssen wrote:

> The PIL manual says that calling image.convert('1') on a greyscale
> image will threshold the image (fairly starkly: "all non-zero values
> are set to 255 (white).")  However, when I try this on a test page of
> color blocks, I seem to get Floyd-Steinberg dithering instead.

> Is the documentation out of date?

most likely.

the draft version at http://effbot.org/imagingbook/image.htm says

    When converting to a bilevel image (mode "1"), the source image
    is first converted to black and white. Resulting values larger than
    127 are then set to white, and the image is dithered. To use other
    thresholds, use the point method.

(but it doesn't mention the dither option, so it needs to be fixed too)

> If I want thresholding, is there a now a way to do it without using
> "point"?  Using quantize(), perhaps?

    im = im.convert("1", dither=Image.NONE)

should work.

> Is there a difference between Image.convert('L') and
> ImageOps.grayscale()?

nope:

    def grayscale(image):
        "Convert to grayscale"
        return image.convert("L")

(all ImageOps operations can be rewritten in terms of core operations;
that module's only purpose is to provide a higher-level interface for some
commonly used operations.  not all operations are quite as trivial as the
one above, though...)

</F> 




From fredrik at pythonware.com  Tue Oct 25 11:07:01 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 25 Oct 2005 11:07:01 +0200
Subject: [Image-SIG] ImageTk.py not installed
References: <435D80A0.9EC24E0B@wadsworth.org>
Message-ID: <djksjs$75l$1@sea.gmane.org>

William Baxter wrote:

> How come ImageTk.py is not put in the site-packages/PIL directory with
> all the other modules? It's there in the
> src/Imaging-1.1.5/build/lib.linux-etc directory when you build Imaging.
> But it doesn't appear in site-packages/PIL during the install. The
> answer is simply to copy it to the place where it's needed, but this has
> been a 'gotcha' everytime I've installed Imaging.

works for me.

what linux release is this, and what Python version ?

(has anyone else seen this?)

</F> 




From janssen at parc.com  Tue Oct 25 19:34:37 2005
From: janssen at parc.com (Bill Janssen)
Date: Tue, 25 Oct 2005 10:34:37 PDT
Subject: [Image-SIG] Image.convert() doesn't dither?
In-Reply-To: Your message of "Tue, 25 Oct 2005 02:04:32 PDT."
	<djksf6$6mm$1@sea.gmane.org> 
Message-ID: <05Oct25.103444pdt."58617"@synergy1.parc.xerox.com>

Thanks!

Bill

From janssen at parc.com  Tue Oct 25 19:44:40 2005
From: janssen at parc.com (Bill Janssen)
Date: Tue, 25 Oct 2005 10:44:40 PDT
Subject: [Image-SIG] Image.convert() doesn't dither?
In-Reply-To: Your message of "Tue, 25 Oct 2005 02:04:32 PDT."
	<djksf6$6mm$1@sea.gmane.org> 
Message-ID: <05Oct25.104444pdt."58617"@synergy1.parc.xerox.com>

>     When converting to a bilevel image (mode "1"), the source image
>     is first converted to black and white. Resulting values larger than

I think it would be clearer to say "the source image is first
converted to grayscale."

>     Resulting values larger than
>     127 are then set to white, and the image is dithered. To use other
>     thresholds, use the point method.

This doesn't make sense to me.  If I take a four-color block, the four
colors having luminance values of 0, 33%, 66%, and 100%, the above
process would give me an image which is half gray (actually, black)
and half white, with some dithering on the gray half.  Instead, I get
four carefully delineated blocks, two of which are dithered gray, one
of which is white, and one of which is black.

I think it makes more sense to say, "the image is dithered to bitonal
black and white using the Floyd-Steinberg error diffusion filter and a
threshold value of 127."

Bill

From s_t_a_n_i at yahoo.com  Tue Oct 25 22:38:47 2005
From: s_t_a_n_i at yahoo.com (www.stani.be)
Date: Tue, 25 Oct 2005 13:38:47 -0700 (PDT)
Subject: [Image-SIG] AGG-based drawing for PIL
In-Reply-To: <djdfeq$mg0$1@sea.gmane.org>
Message-ID: <20051025203847.21147.qmail@web32607.mail.mud.yahoo.com>

Any chance for a cross-platform solution?
Stani
--
SPE - Stani's Python Editor 
http://pythonide.stani.be
http://pythonide.stani.be/manual/html/manual.html

--- Fredrik Lundh <fredrik at pythonware.com> wrote:

> continuing the tradition of announcing new releases
> by replying
> to myself:
> 
> > the "aggdraw" module is a prerelease of a new
> drawing library for PIL,
> > based on Maxim Shemanarev's antigrain (AGG)
> graphics library.  This
> > module implements a WCK-style drawing interface
> (see below), and
> > provides anti-aliasing and alpha compositing.
> 
> A very early 1.2 alpha is now available.
> 
> You can now create "drawable Windows DIB surfaces",
> and blit them
> to screen under any Windows-based toolkit that lets
> you access the
> HWND for a toolkit window.
> 
>     draw = aggdraw.Dib("RGB", (width, height))
>     ... draw as usual ...
>     draw.expose(hwnd) # blit to screen
> 
> Here's a Tkinter example:
> 
>     size = width, height
> 
>     draw = aggdraw.Dib("RGB", size)
>     ... draw as usual ...
> 
>     # display the image
>     frame = Frame(width=size[0], height=size[1],
> bg="")
>     frame.bind("<Expose>", lambda e:
> draw.expose(e.widget.winfo_id()))
>     frame.pack()
> 
> A source kit can be found here:
> 
>     http://effbot.org/downloads/#aggdraw
> 
> To display aggdraw images under Tkinter on other
> platforms, you can use
> PIL's ImageTk.PhotoImage object, or the WCK toolkit.
>  See
> 
>     http://effbot.org/zone/wck-aggview.htm
> 
> for an example on how to use the latter.
> 
> </F>
> 
> 
> 
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig
> 



		
__________________________________ 
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com

From fredrik at pythonware.com  Wed Oct 26 12:16:42 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 26 Oct 2005 12:16:42 +0200
Subject: [Image-SIG] AGG-based drawing for PIL
References: <djdfeq$mg0$1@sea.gmane.org>
	<20051025203847.21147.qmail@web32607.mail.mud.yahoo.com>
Message-ID: <djnl2c$n3r$1@sea.gmane.org>

Stani wrote:

> Any chance for a cross-platform solution?

aggdraw in itself is cross-platform.  if you're not on Windows, or prefer
to write portable code, you can either use tostring() or PIL to transfer
the images from aggdraw's internal buffer to the format used by the UI
toolkit you're using.

for example, you can write portable Tkinter code via PIL's ImageTk layer:

    im = Image.new("RGB", size)
    draw = aggdraw.Draw(im)
    ...
    draw.flush()
    del draw

    image = ImageTk.PhotoImage(im)

or you can use something like the WCK's ui_image method:

    draw = aggdraw.Draw(mode, size)
    ...
    image = self.ui_image(draw.mode, draw.size, draw.tostring())

for wxPython, something like:

    draw = aggdraw.Draw("RGB", size)
    ...
    image = wx.EmptyImage(*size)
    image.SetData(draw.tostring())

should work.

</F>




From arnd.baecker at web.de  Wed Oct 26 13:08:22 2005
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 26 Oct 2005 13:08:22 +0200 (CEST)
Subject: [Image-SIG] AGG-based drawing for PIL
In-Reply-To: <djnl2c$n3r$1@sea.gmane.org>
References: <djdfeq$mg0$1@sea.gmane.org>
	<20051025203847.21147.qmail@web32607.mail.mud.yahoo.com>
	<djnl2c$n3r$1@sea.gmane.org>
Message-ID: <Pine.LNX.4.51.0510261240330.28884@ptpcp8.phy.tu-dresden.de>

On Wed, 26 Oct 2005, Fredrik Lundh wrote:

> for example, you can write portable Tkinter code via PIL's ImageTk layer:
>
>     im = Image.new("RGB", size)
>     draw = aggdraw.Draw(im)
>     ...
>     draw.flush()
>     del draw
>
>     image = ImageTk.PhotoImage(im)
>
> or you can use something like the WCK's ui_image method:
>
>     draw = aggdraw.Draw(mode, size)
>     ...
>     image = self.ui_image(draw.mode, draw.size, draw.tostring())
>
> for wxPython, something like:
>
>     draw = aggdraw.Draw("RGB", size)
>     ...
>     image = wx.EmptyImage(*size)
>     image.SetData(draw.tostring())

Does the ".tostring()" part cost a lot of performance?
I.e. could there be more direct approaches which would be even faster?

(I remember there was some discussion on the matplotlib
mailing list concerning fast conversion of agg buffers to wx, see
http://sourceforge.net/mailarchive/message.php?msg_id=12801822
and the corresponding _wxagg.cpp
http://cvs.sourceforge.net/viewcvs.py/matplotlib/matplotlib/src/_wxagg.cpp?rev=1.1&view=auto
)

Best, Arnd

From Chris.Barker at noaa.gov  Wed Oct 26 20:17:23 2005
From: Chris.Barker at noaa.gov (Chris Barker)
Date: Wed, 26 Oct 2005 11:17:23 -0700
Subject: [Image-SIG] AGG-based drawing for PIL
In-Reply-To: <Pine.LNX.4.51.0510261240330.28884@ptpcp8.phy.tu-dresden.de>
References: <djdfeq$mg0$1@sea.gmane.org>	<20051025203847.21147.qmail@web32607.mail.mud.yahoo.com>	<djnl2c$n3r$1@sea.gmane.org>
	<Pine.LNX.4.51.0510261240330.28884@ptpcp8.phy.tu-dresden.de>
Message-ID: <435FC833.8050409@noaa.gov>

Arnd Baecker wrote:
> Does the ".tostring()" part cost a lot of performance?

Well, I think you get at least two copies of the data: one to make the 
string, and then another one to make the wx.image. It would be nice to 
just pass a pointer to the data buffer right through to wx.Image.

Fredrik,

Have you seen the new "array interface" put out by the NumPy folks?

http://numeric.scipy.org/array_interface.html

It's not intended to by NumPy specific. the idea is to have a standard 
interface so that different Python extensions that need to pass blocks 
of homogeneous data around (like images, for instance) could have a 
standard interface for doing so. That way each extension only needs to 
know about the interface, not each other extension it might need to work 
with. This could be a lot more efficient than using PyStrings to pass 
data around. I'm hoping to implement it in wxPython for just this kind 
of thing. It would be great if PIL did it too. I know a lot of folks do 
work with PIL images and NumPy arrays.

The Python Buffer object is another option.

-Chris

-- 
Christopher Barker, Ph.D.
Oceanographer
                                     		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov

From janssen at parc.com  Wed Oct 26 20:11:14 2005
From: janssen at parc.com (Bill Janssen)
Date: Wed, 26 Oct 2005 11:11:14 PDT
Subject: [Image-SIG] AGG-based drawing for PIL
In-Reply-To: Your message of "Wed, 26 Oct 2005 03:16:42 PDT."
	<djnl2c$n3r$1@sea.gmane.org> 
Message-ID: <05Oct26.111120pdt."58617"@synergy1.parc.xerox.com>

> aggdraw in itself is cross-platform.  if you're not on Windows, or prefer
> to write portable code, you can either use tostring() or PIL to transfer
> the images from aggdraw's internal buffer to the format used by the UI
> toolkit you're using.

Or you could try using pycairo (http://cairographics.org/pycairo)
instead, a portable imaging solution.  Fredrik, are you using
Cairo under WCK?

Cairo interests me because it has the ability to exploit hardware
acceleration on all the major OS platforms.

Bill

From james_hu at hotmail.com  Sun Oct  9 04:57:54 2005
From: james_hu at hotmail.com (James HU)
Date: Sun, 09 Oct 2005 02:57:54 -0000
Subject: [Image-SIG] PIL  Image can open all types of pnf files ?
Message-ID: <BAY7-F21FA43E1D60F0E7AF951FEFD860@phx.gbl>

An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051009/90a09dd7/attachment.htm

From james_hu at hotmail.com  Sun Oct  9 05:20:24 2005
From: james_hu at hotmail.com (James HU)
Date: Sun, 09 Oct 2005 03:20:24 -0000
Subject: [Image-SIG] sorry for posting the same Qs again, but
Message-ID: <BAY7-F254B6E7FFB0E98028E767AFD860@phx.gbl>

An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051009/718c65b4/attachment.html

From andrea_gavana at tin.it  Wed Oct 12 00:16:00 2005
From: andrea_gavana at tin.it (Andrea Gavana)
Date: Tue, 11 Oct 2005 22:16:00 -0000
Subject: [Image-SIG] Question On "Changing" Colour
Message-ID: <00c901c5cf7a$1f615780$69820257@Gavana>

Hello NG,

    I have a small image that looks like a GUI button (with 3D effects given by different pixels colours). The current image has as "basic" colour the grey. For "basic", I mean that the predominant colour in the image is grey and some other pixels are brighter or darker in order to give the 3D effect on that image.

Now my question: is it possible to transform the pixels colours in order to have another basic colour (say blue)? In other words, the predominant colour will become the blue, with other pixels in a brighter or darker blue to give the same 3D effects.

Thanks a lot for every suggestion, and sorry if it is a basic question. I just started to learn PIL.

Andrea.

"Imagination Is The Only Weapon In The War Against Reality."
http://xoomer.virgilio.it/infinity77
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051011/e4aa5edf/attachment.htm

From jhu at metrigenix.com  Sat Oct 15 00:06:06 2005
From: jhu at metrigenix.com (James Hu)
Date: Fri, 14 Oct 2005 22:06:06 -0000
Subject: [Image-SIG] open tiff fil in wxImage directlye
Message-ID: <90DEA7ED93F71A4580CEDF76A02682EA03ACC2@torexch.metrigenix.com>

Hi, all gurus,

 

I need to display 16 bit tiff image on screen directly, but it doesn't
work, only black or gray background shown, (converting from tiff to png
first is not option though). I checked the document, which says wxImage
can open tif! 

Any info or help will be appreciated greatly!!!

 

BTW, I sent this message in another account, but it doesn't seem to be
successful after a few hours.

However, if you guys see this message posted twice, please forgive me.

James

 

image=wx.Image('output.tif',wxBITMAP_TYPE_TIF)   #can't see, still black
as the empty bitmap

 

???do some convertion here? Which function should be used, I couldn't
find one.

 

bmp = image.ConvertToBitmap()

bmp.SetHeight(self.bmpHeight)

bmp.SetWidth(self.bmpWidth)

self.staticBitmap1.SetBitmap(bmp)   #staticBitmap1 already
defined/created before

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051014/6b7c231b/attachment.html

From andrea_gavana at tin.it  Sun Oct 16 23:58:43 2005
From: andrea_gavana at tin.it (Andrea Gavana)
Date: Sun, 16 Oct 2005 21:58:43 -0000
Subject: [Image-SIG] Help On Converting Colours...
Message-ID: <00ab01c5d365$89e12b00$f1a20257@Gavana>

Hello NG,

    I have a small image that looks like a GUI button (with 3D effects given by different pixels colours). The current image has as "basic" colour the grey. For "basic", I mean that the predominant colour in the image is grey and some other pixels are brighter or darker in order to give the 3D effect on that image.

Now my question: is it possible to transform the pixels colours in order to have another basic colour (say blue)? In other words, the predominant colour will become the blue, with other pixels in a brighter or darker blue to give the same 3D effects.

Thanks a lot for every suggestion, and sorry if it is a basic question. I just started to learn PIL.

Andrea.

P.S. I am trying also to use Numarray (Numeric) in order to do this transformation. Someone suggested me to use the function rgb_to_hsv on every pixel and then to change only the "hue" value, to obtain another colour for every pixel, in order to obtain the same 3D effect. 
I think I have found the way to do it, but doing that for every pixel is quite slow... is there a way to do it more "globally" (I mean on the whole image)? Or, as alternative, can I identify the pixels that share the same colour and then apply the transformation on all them?

Thanks a lot for all your suggestions.

Andrea.

"Imagination Is The Only Weapon In The War Against Reality."
http://xoomer.virgilio.it/infinity77
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051016/0870ab03/attachment.htm

From bobspoon at funnylab.co.kr  Mon Oct 31 17:50:47 2005
From: bobspoon at funnylab.co.kr (=?ks_c_5601-1987?B?w9bBpL/t?=)
Date: Tue, 1 Nov 2005 01:50:47 +0900
Subject: [Image-SIG] palette mode animated GIF doesn't show correctly when
	frame > 1
Message-ID: <000001c5de3b$3e2f4850$bb87fbcb@rok>

Python 2.4

PIL 1.1.5

 

 

 

Hi,

I want to show a gif animation which using indexed color model.

I did like this:

 

             Import Image

             im = Image.open(?animation.gif?) 

             try:

                           im.show() # for debug..

                           Im.seek(im.tell()+1) 

             except:

                           pass

 

when ?animation.gif? is in RGB mode, this code work as expected.

But when the gif image is in indexed(palettized) color mode, it does not
show correctly except first frame.

 

I feel this may be caused by something like color shift or palette
corruption ..

 

 

 

 

 

Image is here:

http://203.251.135.188/yoshin/animation.gif

 

 

Best reguard,

             Choi Jung-uk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/image-sig/attachments/20051101/fb44c732/attachment.html