From federo at email.si  Mon Sep  1 15:48:40 2008
From: federo at email.si (Federo)
Date: Mon, 01 Sep 2008 15:48:40 +0200
Subject: [Image-SIG] PyTesser: OCR is not triggered ...
Message-ID: <20080901134841.EBAEE9B250@www1.email.si>

Hi

I am new in PyTesser. I have instolled PyTesser from 
http://code.google.com/p/pytesser/
and PIL from
http://www.pythonware.com/products/pil/


PIL has exe for instalation. For PyTesser I simply unziped all files to
c:\Python25\lib\side-packages

Than I tried the following code:
import Image, ImageGrab
import pytesser
# Taking ScreenShot
im = ImageGrab.grab()
# Konverzija image Formata - Ni nujno potrebna
if im.mode=='RGBA': im=im.convert('RGB')
# PyTesser funkcija izvede OCR
print = image_to_string(im)

I recived the following error:
Traceback (most recent call last):
  File "<interactive input>", line 1, in <module>
  File "C:\Python25\lib\site-packages\pytesser.py", line 31, in image_to_string
    call_tesseract(scratch_image_name, scratch_text_name_root)
  File "C:\Python25\lib\site-packages\pytesser.py", line 21, in call_tesseract
    proc = subprocess.Popen(args)
  File "C:\Python25\lib\subprocess.py", line 594, in __init__
    errread, errwrite)
  File "C:\Python25\lib\subprocess.py", line 816, in _execute_child
    startupinfo)
WindowsError: [Error 2] The system cannot find the file specified

WHAT IS MISSING IN MY INSTALATION?

Loking forward to your reply, Fedo


____________________
http://www.email.si/


From olau at iola.dk  Mon Sep  1 18:26:36 2008
From: olau at iola.dk (Ole Laursen)
Date: Mon, 1 Sep 2008 18:26:36 +0200
Subject: [Image-SIG] Bug in rotate with expand=True
In-Reply-To: <e7dee3cb0804040231s6373589fhc82146915615ce14@mail.gmail.com>
References: <e7dee3cb0804040231s6373589fhc82146915615ce14@mail.gmail.com>
Message-ID: <e7dee3cb0809010926n2a7661e9n9b2deb3bb8054f37@mail.gmail.com>

I hate to quote myself, but I just realized that I had to apply this
patch again against a new server install. Did somebody look at my bug
report from back in April?

It's a one-line fix so it really shouldn't be too hard to review. :)
Is this the wrong forum?

Again, please keep the CC as I'm not subscribed to the list.

On Fri, Apr 4, 2008 at 11:31 AM, Ole Laursen <olau at iola.dk> wrote:
> Hi,
>
> Try this
>
> import Image, ImageDraw
>
> i = Image.new("RGBA", (100, 100), (255, 0, 0, 0))
> draw = ImageDraw.Draw(i)
> draw.line([(0, 50), (100, 50)], fill=(255, 0, 0), width=3)
> i.save("original.png")
>
> i.rotate(15, Image.BICUBIC, True).save("enlarged.png")
> i.rotate(15, Image.BICUBIC, False).save("not-enlarged.png")
>
> Then take a look at "enlarged.png". Contrary to "not-enlarged.png",
> it's not anti-aliased in spite of both having Image.BICUBIC set. I
> believe this is a bug in rotate(). It appears to be a question of line
> 1320 in Image.py
>
>            return self.transform((w, h), AFFINE, matrix)
>
> missing the resample parameter
>
>            return self.transform((w, h), AFFINE, matrix, resample)
>
>
> (Please keep the CC as I'm not subscribed to the list.)
>
> --
> Ole Laursen
> http://www.iola.dk/

-- 
Ole Laursen
http://www.iola.dk/

From fredrik at pythonware.com  Mon Sep  1 18:56:29 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 01 Sep 2008 18:56:29 +0200
Subject: [Image-SIG] Bug in rotate with expand=True
In-Reply-To: <e7dee3cb0809010926n2a7661e9n9b2deb3bb8054f37@mail.gmail.com>
References: <e7dee3cb0804040231s6373589fhc82146915615ce14@mail.gmail.com>
	<e7dee3cb0809010926n2a7661e9n9b2deb3bb8054f37@mail.gmail.com>
Message-ID: <g9h6s0$s47$1@ger.gmane.org>

Ole Laursen wrote:

> I hate to quote myself, but I just realized that I had to apply this
> patch again against a new server install. Did somebody look at my bug
> report from back in April?

there hasn't been a new release since april, but it's fixed in the 
trunk, and you're credited in the CHANGES file, so it'll definitely 
appear in the next release.

</F>


From gareth at spacetaken.net  Tue Sep  2 13:48:33 2008
From: gareth at spacetaken.net (Gareth)
Date: Tue, 02 Sep 2008 12:48:33 +0100
Subject: [Image-SIG] is this a bug?
Message-ID: <48BD2811.70509@spacetaken.net>

I believe I have found a bug with PIL. It can be illustrated by the 
following code:

from PIL import Image, ImageFile

path = <path to a valid png file>

file = open(path, 'rb')
p = ImageFile.Parser()
p.feed(file.read())
img = p.close()
file.close()

img.thumbnail((380, 380), Image.ANTIALIAS)
img.save(<path in an existing directory>)

- this works fine it the path is not to a png but to a jpg or gif, but 
given that is points to a (valid) png it chucks the following error:

Traceback (most recent call last):
  File "piltest.py", line 12, in <module>
    img.save('C:/Documents and Settings/gareth0/Desktop/thumb.png')
  File "C:\Python25\Lib\site-packages\PIL\Image.py", line 1372, in save
    self.load()
  File "C:\Python25\Lib\site-packages\PIL\ImageFile.py", line 189, in load
    s = read(self.decodermaxblock)
  File "C:\Python25\Lib\site-packages\PIL\PngImagePlugin.py", line 365, 
in load_read
    return self.fp.read(bytes)
  File "C:\Python25\Lib\site-packages\PIL\ImageFile.py", line 300, in read
    data = self.data[pos:pos+bytes]
TypeError: 'NoneType' object is unsubscriptable

- i hope someone can help me out here, this has been driving me insane, 
thanks!


From ashishbitsgoa at gmail.com  Wed Sep  3 11:10:03 2008
From: ashishbitsgoa at gmail.com (Ashish Sethi)
Date: Wed, 3 Sep 2008 14:40:03 +0530
Subject: [Image-SIG] Converting a binary file to ascii
Message-ID: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>

I have PPM-P6 image file which is in binary format .The file is of the type  :-

P6
128 128
255
?????{??~??x??y??f??l??y??j??h??u??y??v??p???ub?^Y?f](c)g ..........and so on.

How can i read the 4th line (binary) and convert it to an equivalent
ascii string. I used the following method but couldn't get the desired
result..

import binascii
from binascii import b2a_uu
infile=open("C:\\Documents and Settings\\Desktop\\lena.ppm")
a=infile.readline()
b=infile.readline()
c=infile.readline()
d=infile.readline()
z=b2a_uu(d)
print z

This script gave me some random absurd data as output.
Plz help

From karsten.hiddemann at mathematik.uni-dortmund.de  Wed Sep  3 11:09:11 2008
From: karsten.hiddemann at mathematik.uni-dortmund.de (Karsten Hiddemann)
Date: Wed, 03 Sep 2008 11:09:11 +0200
Subject: [Image-SIG] Converting a binary file to ascii
In-Reply-To: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
References: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
Message-ID: <48BE5437.8080805@mathematik.uni-dortmund.de>

Ashish Sethi schrieb:
> How can i read the 4th line (binary) and convert it to an equivalent
> ascii string. [...]
> 
> This script gave me some random absurd data as output.

Have you tried using d=infile.read() to really read in *all* the 
remaining data from the file?

From ned at nedbatchelder.com  Wed Sep  3 13:27:43 2008
From: ned at nedbatchelder.com (Ned Batchelder)
Date: Wed, 03 Sep 2008 07:27:43 -0400
Subject: [Image-SIG] Converting a binary file to ascii
In-Reply-To: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
References: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
Message-ID: <48BE74AF.9000007@nedbatchelder.com>

Since you are on Windows, you need to provide the "rb" mode to open, or 
it will convert newline characters in your file, which can trash your 
binary data.  Also, I don't understand what b2a_uu does, but the docs 
say its input string should be at most 45 characters, so it doesn't seem 
like what you want.

When you say "convert it to an equivalent ascii string", what do you mean?

--Ned.
http://nedbatchelder.com

Ashish Sethi wrote:
> I have PPM-P6 image file which is in binary format .The file is of the type  :-
>
> P6
> 128 128
> 255
> ?????{??~??x??y??f??l??y??j??h??u??y??v??p???ub?^Y?f](c)g ..........and so on.
>
> How can i read the 4th line (binary) and convert it to an equivalent
> ascii string. I used the following method but couldn't get the desired
> result..
>
> import binascii
> from binascii import b2a_uu
> infile=open("C:\\Documents and Settings\\Desktop\\lena.ppm")
> a=infile.readline()
> b=infile.readline()
> c=infile.readline()
> d=infile.readline()
> z=b2a_uu(d)
> print z
>
> This script gave me some random absurd data as output.
> Plz help
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig
>   

-- 
Ned Batchelder, http://nedbatchelder.com



From adeuring at gmx.net  Wed Sep  3 15:52:02 2008
From: adeuring at gmx.net (abel deuring)
Date: Wed, 03 Sep 2008 15:52:02 +0200
Subject: [Image-SIG] Converting a binary file to ascii
In-Reply-To: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
References: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
Message-ID: <48BE9682.1080500@gmx.net>

On 03.09.2008 11:10, Ashish Sethi wrote:
> I have PPM-P6 image file which is in binary format .The file is of the type  :-
> 
> P6
> 128 128
> 255
> ?????{??~??x??y??f??l??y??j??h??u??y??v??p???ub?^Y?f](c)g ..........and so on.
> 
> How can i read the 4th line (binary) and convert it to an equivalent
> ascii string. I used the following method but couldn't get the desired
> result..

Please define more precisely what you mean with "equivalent ascii
string". Strictly speaking, b2a_uu produces an equivalent string in the
sense that running a2b_uu gives you the original string back, but this
is obviously not what you want. In other words, there is 1:1 mapping
between arbitrary binary data and their "UU" representation.

If you want to get the "P3" format: Use something like

intvals = [ord(i) for i in d]

(where d is the string with the binary image data read from the
P6-format file) and create an ASCII representation for the elements of
the list intvals.

> 
> import binascii
> from binascii import b2a_uu
> infile=open("C:\\Documents and Settings\\Desktop\\lena.ppm")
> a=infile.readline()
> b=infile.readline()
> c=infile.readline()
> d=infile.readline()
> z=b2a_uu(d)
> print z
> 
> This script gave me some random absurd data as output.

As Ned already wrote, you should open the file in binary mode on
Windows. And as Karsten wrote, you should use infile.read() to get the
image data. readline() is not useful to read binary data: You will get
the data up to the first byte with the value 10 (ASCII meaning: "line
feed"), or perhaps for Windows up to the first byte with the value 13
(ASCII meaning "carriage return"). But these values have for P6 files a
different "meaning", when they appear after the leading "255" line: They
simply give the value of the red, blue or green component of a pixel.

Abel

From cannon.el at gmail.com  Wed Sep  3 17:17:03 2008
From: cannon.el at gmail.com (Laura & Edward Cannon)
Date: Wed, 3 Sep 2008 08:17:03 -0700
Subject: [Image-SIG]  is this a bug?
In-Reply-To: <b986b290809030816q696e698fie8d73a80316ae82b@mail.gmail.com>
References: <48BD2811.70509@spacetaken.net>
	<b986b290809030816q696e698fie8d73a80316ae82b@mail.gmail.com>
Message-ID: <b986b290809030817m297fb430v22d047ce2e120f9a@mail.gmail.com>

From: Laura & Edward Cannon <cannon.el at gmail.com>
Date: Wed, Sep 3, 2008 at 8:16 AM
Subject: Re: [Image-SIG] is this a bug?
To: Gareth <gareth at spacetaken.net>


I tried this out as well and got the same error. I agree this seems to
be a bug, but in the short term try to get around it by using the
Image.open() function. code would look like this

import Image
path = <path to a valid png file>
img = Image.open(path)
img.thumbnail((380, 380), Image.ANTIALIAS)
img.save(<path in an existing directory>)

this is probably the preferred method when working with real files,
the ImageFile is mostly for things sent over the network. It is
possible you are also using a png with some interesting compression,
the documentation states "If the file cannot be identified, the parser
will raise an IOError exception in the close method. If the file can
be identified, but not decoded (for example, if the data is damaged,
or if it uses an unsupported compression method), the parser will
raise an IOError exception as soon as possible, either in feed or
close. "

On Tue, Sep 2, 2008 at 4:48 AM, Gareth <gareth at spacetaken.net> wrote:
>
> I believe I have found a bug with PIL. It can be illustrated by the following code:
>
> from PIL import Image, ImageFile
>
> path = <path to a valid png file>
>
> file = open(path, 'rb')
> p = ImageFile.Parser()
> p.feed(file.read())
> img = p.close()
> file.close()
>
> img.thumbnail((380, 380), Image.ANTIALIAS)
> img.save(<path in an existing directory>)
>
> - this works fine it the path is not to a png but to a jpg or gif, but given that is points to a (valid) png it chucks the following error:
>
> Traceback (most recent call last):
>  File "piltest.py", line 12, in <module>
>   img.save('C:/Documents and Settings/gareth0/Desktop/thumb.png')
>  File "C:\Python25\Lib\site-packages\PIL\Image.py", line 1372, in save
>   self.load()
>  File "C:\Python25\Lib\site-packages\PIL\ImageFile.py", line 189, in load
>   s = read(self.decodermaxblock)
>  File "C:\Python25\Lib\site-packages\PIL\PngImagePlugin.py", line 365, in load_read
>   return self.fp.read(bytes)
>  File "C:\Python25\Lib\site-packages\PIL\ImageFile.py", line 300, in read
>   data = self.data[pos:pos+bytes]
> TypeError: 'NoneType' object is unsubscriptable
>
> - i hope someone can help me out here, this has been driving me insane, thanks!
>
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig

From ashishbitsgoa at gmail.com  Thu Sep  4 06:32:27 2008
From: ashishbitsgoa at gmail.com (Ashish Sethi)
Date: Thu, 4 Sep 2008 10:02:27 +0530
Subject: [Image-SIG] Converting a binary file to ascii
In-Reply-To: <48BE9682.1080500@gmx.net>
References: <edb5c69c0809030210y42a64c99tb6ccca58c54b5731@mail.gmail.com>
	<48BE9682.1080500@gmx.net>
Message-ID: <edb5c69c0809032132i4d9c71fflff5e66e2c1193f68@mail.gmail.com>

> Please define more precisely what you mean with "equivalent ascii
> string". Strictly speaking, b2a_uu produces an equivalent string in the
> sense that running a2b_uu gives you the original string back, but this
> is obviously not what you want. In other words, there is 1:1 mapping
> between arbitrary binary data and their "UU" representation.
>
> If you want to get the "P3" format: Use something like
>
> intvals = [ord(i) for i in d]

Thanks all of you. Your suggestions solved my problem.

From fredrik at pythonware.com  Sat Sep  6 13:18:16 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 06 Sep 2008 13:18:16 +0200
Subject: [Image-SIG] is this a bug?
In-Reply-To: <48BD2811.70509@spacetaken.net>
References: <48BD2811.70509@spacetaken.net>
Message-ID: <g9totp$ibq$1@ger.gmane.org>

Gareth wrote:

> I believe I have found a bug with PIL. It can be illustrated by the 
> following code:
> 
> from PIL import Image, ImageFile
> 
> path = <path to a valid png file>
> 
> file = open(path, 'rb')
> p = ImageFile.Parser()
> p.feed(file.read())
> img = p.close()
> file.close()
> 
> img.thumbnail((380, 380), Image.ANTIALIAS)
> img.save(<path in an existing directory>)
> 
> - this works fine it the path is not to a png but to a jpg or gif, but 
> given that is points to a (valid) png it chucks the following error:

> TypeError: 'NoneType' object is unsubscriptable

sure looks like a bug to me.  as a workaround, try changing the "except 
IOError:" line in this section of ImageFile.py

             # attempt to open this file
             try:
                 try:
                     fp = _ParserFile(self.data)
                     im = Image.open(fp)
                 finally:
                     fp.close() # explicitly close the virtual file
             except IOError: # <-- this line
                 pass # not enough data

to

             except (IOError, TypeError):

also, if you can, mail me a copy of the offending file (or post it 
somewhere and send me an URL) so I can investigate further.

</F>


From david at flowgram.com  Sat Sep  6 02:24:16 2008
From: david at flowgram.com (David Gobaud)
Date: Fri, 5 Sep 2008 17:24:16 -0700
Subject: [Image-SIG] Image paste method resulting in garbage pixels in images
Message-ID: <028c01c90fb6$e5fb83a0$fdfefe0a@stanford.edu>

Hi,
 
We are using PIL to render images. When we use the Image paste method to
blend 2 RGBA image we end up with garbage pixels in the output as shown
below. How can we fix this?

Thank you,


David Gobaud




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080905/c54f3715/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 24791 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080905/c54f3715/attachment-0001.jpeg>

From fredrik at pythonware.com  Sun Sep  7 11:12:23 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 07 Sep 2008 11:12:23 +0200
Subject: [Image-SIG] Image paste method resulting in garbage pixels in
	images
In-Reply-To: <028c01c90fb6$e5fb83a0$fdfefe0a@stanford.edu>
References: <028c01c90fb6$e5fb83a0$fdfefe0a@stanford.edu>
Message-ID: <ga05to$rv6$1@ger.gmane.org>

David Gobaud wrote:

> We are using PIL to render images. When we use the Image paste method to 
> blend 2 RGBA image we end up with garbage pixels in the output as shown 
> below. How can we fix this?

can you send me the two image layers used to create that image? (or post 
them somewhere and mail me a link)

</F>


From snaury at gmail.com  Sun Sep  7 13:22:05 2008
From: snaury at gmail.com (Alexey Borzenkov)
Date: Sun, 7 Sep 2008 15:22:05 +0400
Subject: [Image-SIG] Announcing my PIL repository
Message-ID: <e2480c70809070422u77159614kdf340b16e6833b4a@mail.gmail.com>

Hello, fellow Image-SIG subscribers,

PIL 1.1.6 has been released almost two years ago. And as some of you
may remember, I have sent several patches to this list since then, but
there hasn't been any reply to my emails from developers for some
reason, and I don't even know if my patches are accepted or not, if
they require any work, etc.

Anyway, previously my PIL branch was part of a bigger repository that
I couldn't open to everyone, but now I have found a way to split it
into a separate repository (using bzr-svn), so everyone can look at it
if they want to:

    svn://kitsu.ru/pil/trunk

Many revisions wouldn't be of any interest to anybody (they just
fiddle with setup scripts), but there are some features (I haven't
sent to this list yet) that might be interesting to some of you who do
raw image data processing and use Image.fromstring and Image.tostring
heavily:

    - Add P;4R raw type, where high and low 4-bits are swapped.
    - Added RGB to RGBa dummy conversion, plus added BGR;15 and BGR;16
packing code.
    - Added a very simple RGBa->RGBA conversion

And of course it includes all the patches I sent to this list.

Binaries for Python 2.4 and Python 2.5 on Windows are also available:

    http://kitsu.ru/pil

The are built using mingw (gcc 4.3.0) and are linked against jpeg-6b,
tiff-3.8.2, freetype-2.3.7 and ActiveTcl 8.4.19.1.

P.S. Please let me know if there is anything else missing.
P.P.S. And I hope that PIL is not dead and we will see some releases
in the future. Almost two years without any new release and without
any public repository to watch for is very tough. :(

From snaury at gmail.com  Sun Sep  7 16:53:04 2008
From: snaury at gmail.com (Alexey Borzenkov)
Date: Sun, 7 Sep 2008 18:53:04 +0400
Subject: [Image-SIG] Announcing my PIL repository
In-Reply-To: <48C3DA28.9060304@users.sourceforge.net>
References: <e2480c70809070422u77159614kdf340b16e6833b4a@mail.gmail.com>
	<48C3DA28.9060304@users.sourceforge.net>
Message-ID: <e2480c70809070753q1d5765aex3bccd0c3212c06d9@mail.gmail.com>

On Sun, Sep 7, 2008 at 5:42 PM, Pander <pander at users.sourceforge.net> wrote:
> Thanks for your work. Did you get any replies from the original authors?

Well, It's just that nobody replied to my patch emails. They went to
the list completely silently, and that felt a little bit odd. :-/

> Attached you will find a contribution from my side. I hope it can find
> its way into PIL. Please review the code and report back any issues you
> might encounter.

I'm afraid you might have misunderstood me. I don't have anything to
do with PythonWare or Secret Labs AB, so I absolutely can't decide
what goes and what doesn't go into PIL. :-/

As for the review I see at least some problems with your code:

1) It's too specialized. I would recommend, if possible, splitting
image generation and caching code into different functions, so that
one returns unsaved images and the other checks file cache and in case
of a cache-miss uses the first one and saves it on disk.

2) You hard-code text to be str in utf-8 encoding. Even worse,
different functions expect different things (for example,
getWidth[Height] expects a ascii/unicode string, but generateLabel
expects utf-8 string), you even make the mistake of incorrectly
calling them in your own code. Just make users call your functions
with ascii/unicode strings and convert to unicode if you really need
to, don't special-case on utf-8 or any other encoding.

~ Alexey

From fredrik at pythonware.com  Sun Sep  7 17:07:00 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 07 Sep 2008 17:07:00 +0200
Subject: [Image-SIG] Announcing my PIL repository
In-Reply-To: <e2480c70809070422u77159614kdf340b16e6833b4a@mail.gmail.com>
References: <e2480c70809070422u77159614kdf340b16e6833b4a@mail.gmail.com>
Message-ID: <ga0qml$18f$1@ger.gmane.org>

Alexey Borzenkov wrote:

> P.P.S. And I hope that PIL is not dead and we will see some releases
> in the future.

http://mail.python.org/pipermail/image-sig/2008-July/005101.html

</F>


From agi20dla at gmail.com  Mon Sep  8 01:12:45 2008
From: agi20dla at gmail.com (agi20dla)
Date: Sun, 07 Sep 2008 19:12:45 -0400
Subject: [Image-SIG] IP Geolocation
Message-ID: <48C45FED.1050806@gmail.com>

I created an IP geolocation script with another popular scripting
language using the Tiger/LINE database from the Census Bureau.  Is there
such a module for Python?  I need to get street level IP Geo for at
least USA addresses, but also for overseas addresses if possible.

agi20dla

From jwevandijk at xs4all.nl  Mon Sep  8 16:17:41 2008
From: jwevandijk at xs4all.nl (Janwillem van Dijk)
Date: Mon, 08 Sep 2008 16:17:41 +0200
Subject: [Image-SIG] Image.open problem
Message-ID: <48C53405.7090805@xs4all.nl>

It all worked nicely but after a re-install of my systems (linux pc, 
linux laptop, windows pc) I have a problem with Image.open(fname). Under 
linux (python2.5 ubuntu 8.04) I get:

Traceback (most recent call last):
  File "/media/DATAWINDOWS/Python/SobelQuality/open_image.py", line 6, 
in <module>
    image = Image.open(fname)
  File "/usr/lib/python2.5/site-packages/PIL/Image.py", line 1902, in open
    return factory(fp, filename)
  File "/usr/lib/python2.5/site-packages/PIL/ImageFile.py", line 82, in 
__init__
    self._open()
  File "/usr/lib/python2.5/site-packages/PIL/JpegImagePlugin.py", line 
270, in _open
    handler(self, i)
  File "/usr/lib/python2.5/site-packages/PIL/JpegImagePlugin.py", line 
154, in DQT
    self.quantization[v&15] = array.array("b", s[1:65])
AttributeError: 'module' object has no attribute 'array'

with this simple test script:
#open test image
import sys, os.path, time
from PIL import Image
t = time.time()
fname = os.path.join(sys.path[0],'lena.jpg')
image = Image.open(fname)
print 'Open %s width,height %d,%d took %.3f sec' % \
    (fname,image.size[0],image.size[1],time.time()-t)

On windows XP it works as it should.

What went wrong and how to get it running again

many thanks,
Janwillem



From fredrik at pythonware.com  Mon Sep  8 16:49:27 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 08 Sep 2008 16:49:27 +0200
Subject: [Image-SIG] Image.open problem
In-Reply-To: <48C53405.7090805@xs4all.nl>
References: <48C53405.7090805@xs4all.nl>
Message-ID: <ga3e1p$19m$1@ger.gmane.org>

Janwillem van Dijk wrote:

> It all worked nicely but after a re-install of my systems (linux pc, 
> linux laptop, windows pc) I have a problem with Image.open(fname). Under 
> linux (python2.5 ubuntu 8.04) I get:
> 
> Traceback (most recent call last):
>  File "/media/DATAWINDOWS/Python/SobelQuality/open_image.py", line 6, in 
> <module>
>    image = Image.open(fname)
>  File "/usr/lib/python2.5/site-packages/PIL/Image.py", line 1902, in open
>    return factory(fp, filename)
>  File "/usr/lib/python2.5/site-packages/PIL/ImageFile.py", line 82, in 
> __init__
>    self._open()
>  File "/usr/lib/python2.5/site-packages/PIL/JpegImagePlugin.py", line 
> 270, in _open
>    handler(self, i)
>  File "/usr/lib/python2.5/site-packages/PIL/JpegImagePlugin.py", line 
> 154, in DQT
>    self.quantization[v&15] = array.array("b", s[1:65])
> AttributeError: 'module' object has no attribute 'array'
> 
> with this simple test script:
> #open test image
> import sys, os.path, time
> from PIL import Image
> t = time.time()
> fname = os.path.join(sys.path[0],'lena.jpg')
> image = Image.open(fname)
> print 'Open %s width,height %d,%d took %.3f sec' % \
>    (fname,image.size[0],image.size[1],time.time()-t)

PIL's JPEG module attempts to load the standard library module called 
"array", but it looks as if "import array" fetches some other module.

check your Python paths for things named "array.py" or similar.

</F>


From cannon.el at gmail.com  Mon Sep  8 20:04:00 2008
From: cannon.el at gmail.com (Laura & Edward Cannon)
Date: Mon, 8 Sep 2008 11:04:00 -0700
Subject: [Image-SIG] IP Geolocation
In-Reply-To: <48C45FED.1050806@gmail.com>
References: <48C45FED.1050806@gmail.com>
Message-ID: <b986b290809081104y16047ba5la6acb8bff620c16d@mail.gmail.com>

On Sun, Sep 7, 2008 at 4:12 PM, agi20dla <agi20dla at gmail.com> wrote:
> I created an IP geolocation script with another popular scripting
> language using the Tiger/LINE database from the Census Bureau.  Is there
> such a module for Python?  I need to get street level IP Geo for at
> least USA addresses, but also for overseas addresses if possible.
>
> agi20dla
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig
>
It sounds like you need a commercial solution, this kind of thing is
not at all what PIL does, PIL is a general purpose image library.

From snaury at gmail.com  Mon Sep  8 20:54:50 2008
From: snaury at gmail.com (Alexey Borzenkov)
Date: Mon, 8 Sep 2008 22:54:50 +0400
Subject: [Image-SIG] Announcing my PIL repository
In-Reply-To: <ca816810809080628w570e0677i7f958a9ea2d86b2f@mail.gmail.com>
References: <e2480c70809070422u77159614kdf340b16e6833b4a@mail.gmail.com>
	<48C3DA28.9060304@users.sourceforge.net>
	<e2480c70809070753q1d5765aex3bccd0c3212c06d9@mail.gmail.com>
	<ca816810809071801s752c39e1u213ccd0dcd65f608@mail.gmail.com>
	<e2480c70809072132y789616d6k5c32fa990553979c@mail.gmail.com>
	<e2480c70809072140m70dabfb7h35d1393723b74577@mail.gmail.com>
	<ca816810809080628w570e0677i7f958a9ea2d86b2f@mail.gmail.com>
Message-ID: <e2480c70809081154t17436277n1221d583c837ca98@mail.gmail.com>

On Mon, Sep 8, 2008 at 5:28 PM, Kent Tenney <ktenney at gmail.com> wrote:
>> Ah, I found what you are talking about.
>>
>> There is a patch by Lowell Alleman, who posted it to this list on 23
>> February 2007, and it's a pure-python patch. So, no, this was not my
>> patch. :)
> +1 to preserving PngInfo by default.

Why would you want to do so by default? If you want to have text
chunks (or any other chunks) you can use PngInfo yourself:

pnginfo = PngInfo()
pnginfo.add_text("blah", "some value")
pnginfo.add_text("another", "some other value", zip=1)
im.save("filename.png", pnginfo=pnginfo)

Other than that text chunks are currently exposed via im_info
dictionary, which cannot be used for generating tEXt chunks as it is
(this dictionary has other keys that have nothing to do with text
chunks, and any backwards compatible solution would have to either be
guesswork, which is -1 by me, or a separate set/dictionary containing
a set of keys that are text chunks, and maybe a flag whether it was
tEXt or zTXt, which is not ideal either).

In what situations do you find it useful so preserve text tags by default?

From agi20dla at gmail.com  Tue Sep  9 12:59:12 2008
From: agi20dla at gmail.com (Jason S)
Date: Tue, 09 Sep 2008 06:59:12 -0400
Subject: [Image-SIG] IP Geolocation
In-Reply-To: <b986b290809081104y16047ba5la6acb8bff620c16d@mail.gmail.com>
References: <48C45FED.1050806@gmail.com>
	<b986b290809081104y16047ba5la6acb8bff620c16d@mail.gmail.com>
Message-ID: <48C65700.8090905@gmail.com>

Laura & Edward Cannon wrote:
> On Sun, Sep 7, 2008 at 4:12 PM, agi20dla <agi20dla at gmail.com> wrote:
>   
>> I created an IP geolocation script with another popular scripting
>> language using the Tiger/LINE database from the Census Bureau.  Is there
>> such a module for Python?  I need to get street level IP Geo for at
>> least USA addresses, but also for overseas addresses if possible.
>>
>> agi20dla
>> _______________________________________________
>> Image-SIG maillist  -  Image-SIG at python.org
>> http://mail.python.org/mailman/listinfo/image-sig
>>
>>     
> It sounds like you need a commercial solution, this kind of thing is
> not at all what PIL does, PIL is a general purpose image library.
>
>   
I posted here because the Image SIG said this would provide

"...an open forum for Python users interested in image handling and
processing; anything from simply reading and writing image files to
scientific image processing, *****geographical information***** systems,
remote sensing, and more. "

If that's not IP Geolocation, do you know of a SIG that can help me? The
commercial solutions are not that good or are expensive.

J

From newz at bearfruit.org  Tue Sep  9 16:08:30 2008
From: newz at bearfruit.org (Matthew Nuzum)
Date: Tue, 9 Sep 2008 09:08:30 -0500
Subject: [Image-SIG] IP Geolocation
In-Reply-To: <48C65700.8090905@gmail.com>
References: <48C45FED.1050806@gmail.com>
	<b986b290809081104y16047ba5la6acb8bff620c16d@mail.gmail.com>
	<48C65700.8090905@gmail.com>
Message-ID: <f3c0b4080809090708j8754db9ldb872b15b614e5ef@mail.gmail.com>

On Tue, Sep 9, 2008 at 5:59 AM, Jason S <agi20dla at gmail.com> wrote:
> Laura & Edward Cannon wrote:
>> On Sun, Sep 7, 2008 at 4:12 PM, agi20dla <agi20dla at gmail.com> wrote:
>>> I created an IP geolocation script with another popular scripting
>>> language using the Tiger/LINE database from the Census Bureau.  Is there
>>> such a module for Python?  I need to get street level IP Geo for at
>>> least USA addresses, but also for overseas addresses if possible.
>>>
>> It sounds like you need a commercial solution, this kind of thing is
>> not at all what PIL does, PIL is a general purpose image library.

> I posted here because the Image SIG said this would provide

Well, let me just say this, having been listening on this list for
quite a while there has not been much discussion on this (I can't
remember any at least).

Therefore if you're willing to be a bit of a pioneer this is probably
a fine place to discuss it.

I see two parts:

 1. getting the geoip info out of the image
 2. doing the lat/long to address lookup

Part 1 is definitely on-topic, though I can be of little help. I don't
see any reason why you'd have to pay for a tool to extract the data to
do this (though sometimes that does help you get your task done more
quickly).

Regarding part 2, I've done address to lat/long lookup using google's
apis but the reverse is not possible directly (though there is some
community contributed tools for doing it, with limitations). However
you may get more experienced and knowledgable answers on this last
part in other forums. I've done my research in the context of Google
maps and found some good information in their google group.

Back to part 1, is this information stored in the exif header? Do you
have a sample image you can post somewhere?
-- 
Matthew Nuzum
newz2000 on freenode

From fredrik at pythonware.com  Fri Sep 12 17:51:01 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 12 Sep 2008 17:51:01 +0200
Subject: [Image-SIG] pil sightings
Message-ID: <gae358$74h$1@ger.gmane.org>

this definitely made my day:

   http://vanrees.org/weblog/archive/2008/09/12/python-calculated-coin

excellent work, stani.  and congrats for winning!

(just wish I could figure out how to order one from here ;-)

</F>


From Bob.Klepfer at dynetics.com  Fri Sep 12 18:53:18 2008
From: Bob.Klepfer at dynetics.com (Klepfer, Bob)
Date: Fri, 12 Sep 2008 11:53:18 -0500
Subject: [Image-SIG] pil sightings
In-Reply-To: <gae358$74h$1@ger.gmane.org>
References: <gae358$74h$1@ger.gmane.org>
Message-ID: <C26858FE3896984B94AB3D2B0C834B3A7A8430@hyperion.in.dynetics.com>

This is brilliant!  Anyone from the list willing to arrange to mail one
over the pond?

Bob

--
Bob Klepfer
Research Analyst
Dynetics, Inc.
Huntsville, AL  35806
http://dynetics.com


-----Original Message-----
From: image-sig-bounces+bob.klepfer=dynetics.com at python.org
[mailto:image-sig-bounces+bob.klepfer=dynetics.com at python.org] On Behalf
Of Fredrik Lundh
Sent: Friday, September 12, 2008 10:51 AM
To: image-sig at python.org
Subject: [Image-SIG] pil sightings

this definitely made my day:

   http://vanrees.org/weblog/archive/2008/09/12/python-calculated-coin

excellent work, stani.  and congrats for winning!

(just wish I could figure out how to order one from here ;-)

</F>

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

From pricewhelan at mac.com  Sat Sep 13 07:26:21 2008
From: pricewhelan at mac.com (Adrian Price-Whelan)
Date: Sat, 13 Sep 2008 01:26:21 -0400
Subject: [Image-SIG] PIL Compile Error Mac 10.5.4
Message-ID: <E756702C-91AC-48B6-B40C-5169576607DA@mac.com>

Hey -

Having an issue building PIL on my intel Macbook Pro running 10.5.4.:

ld: in /sw/lib/libxml2.2.dylib, file is not of required architecture  
for architecture ppc
collect2: ld returned 1 exit status
lipo: can't open input file: /var/folders/U8/U8oYQjnOFIi8el0mQQiHaU++ 
+TI/-Tmp-//ccZZXwfx.out (No such file or directory)
error: command 'gcc' failed with exit status 1

Not quite sure whats going on with this...any help would be much  
appreciated. Thanks!

-Adrian

From spe.stani.be at gmail.com  Sun Sep 14 22:56:25 2008
From: spe.stani.be at gmail.com (Stani)
Date: Sun, 14 Sep 2008 22:56:25 +0200
Subject: [Image-SIG] pil sightings
In-Reply-To: <gae358$74h$1@ger.gmane.org>
References: <gae358$74h$1@ger.gmane.org>
Message-ID: <1221425785.14933.44.camel@blue>

Hi Fredrik,

At the moment I am working hard with the Royal Dutch Mint to produce the
coin. It is quite fascinating to see money making from the secret
inside. 

When the stress is over, I might do a blog post about the whole
experience and some higher resolution images, as the current ones are
lacking details. The coin is for 90% made with python, PIL & pycairo.
Some touches were done with Gimp & Inkscape. So no one should doubt any
longer that you can make money with free software ;-)

Fredrik, don't worry about ordering a coin, if you email me your postal
address, I'll happily ship one to you. Most of my work as a visual
artist is PIL powered, so I owe you that one ;-)

Other people will be able to order it from the coinshop of the Royal
Dutch Mint (link at bottom), but probably it will be sold out quite
fast:
http://en.knm.nl/domains/knm/pages/article.asp?content=11720000000909_999_11650000000011&navid=11650000000014_11650000000009

Best regards,
Stani

PS For the curious, the following link is the reason why I started
learning PIL in 2002. It's a cross-over between 'The Birds' of Hitchcock
and Vanitas paintings:

http://88.191.26.71/maria/openlayers.html

(Those of you familiar with Chernobyl might recognize the silhouet.)

Op vrijdag 12-09-2008 om 17:51 uur [tijdzone +0200], schreef Fredrik
Lundh:
> this definitely made my day:
> 
>    http://vanrees.org/weblog/archive/2008/09/12/python-calculated-coin
> 
> excellent work, stani.  and congrats for winning!
> 
> (just wish I could figure out how to order one from here ;-)
> 
> </F>
> 
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig


From gwidion at mpc.com.br  Sun Sep 14 23:31:04 2008
From: gwidion at mpc.com.br (Joao S. O. Bueno)
Date: Sun, 14 Sep 2008 18:31:04 -0300
Subject: [Image-SIG] pil sightings
In-Reply-To: <1221425785.14933.44.camel@blue>
References: <gae358$74h$1@ger.gmane.org> <1221425785.14933.44.camel@blue>
Message-ID: <200809141831.04844.gwidion@mpc.com.br>

Hi Stani!

Nice job!!  :-)

I had forwarded the news on the coin to the Brazillian python mailing list, 
and people got quite amused tehre as well.

BTW, this week Brazilian's PyCon will take place,  if you want I think could 
display some coin images/other data as a lightning talk on your behalf.

Great to hear from you on this unsuspected way.  :-)

Since next years' LGM is being scheduled to take place in Amsterdam, I guess I 
will meet you there again.

	js
	-><-

On Sunday 14 September 2008, Stani wrote:
> Hi Fredrik,
>
> At the moment I am working hard with the Royal Dutch Mint to produce the
> coin. It is quite fascinating to see money making from the secret
> inside.
>
> When the stress is over, I might do a blog post about the whole
> experience and some higher resolution images, as the current ones are
> lacking details. The coin is for 90% made with python, PIL & pycairo.
> Some touches were done with Gimp & Inkscape. So no one should doubt any
> longer that you can make money with free software ;-)
>
> Fredrik, don't worry about ordering a coin, if you email me your postal
> address, I'll happily ship one to you. Most of my work as a visual
> artist is PIL powered, so I owe you that one ;-)
>
> Other people will be able to order it from the coinshop of the Royal
> Dutch Mint (link at bottom), but probably it will be sold out quite
> fast:
> http://en.knm.nl/domains/knm/pages/article.asp?content=11720000000909_999_1
>1650000000011&navid=11650000000014_11650000000009
>
> Best regards,
> Stani
>
> PS For the curious, the following link is the reason why I started
> learning PIL in 2002. It's a cross-over between 'The Birds' of Hitchcock
> and Vanitas paintings:
>
> http://88.191.26.71/maria/openlayers.html
>
> (Those of you familiar with Chernobyl might recognize the silhouet.)
>
> Op vrijdag 12-09-2008 om 17:51 uur [tijdzone +0200], schreef Fredrik
>
> Lundh:
> > this definitely made my day:
> >
> >    http://vanrees.org/weblog/archive/2008/09/12/python-calculated-coin
> >
> > excellent work, stani.  and congrats for winning!
> >
> > (just wish I could figure out how to order one from here ;-)
> >
> > </F>
> >
> > _______________________________________________
> > Image-SIG maillist  -  Image-SIG at python.org
> > http://mail.python.org/mailman/listinfo/image-sig
>
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig



From oliver.siemoneit at philo.uni-stuttgart.de  Fri Sep 12 18:30:53 2008
From: oliver.siemoneit at philo.uni-stuttgart.de (Oliver Siemoneit)
Date: Fri, 12 Sep 2008 18:30:53 +0200
Subject: [Image-SIG] Image Correction For Colorblind Users
Message-ID: <6B8DFA6E1AD34EA5994FB887040BDFB8@dusk>

Hi,

I have implemented two image correction algorithms which probably could make 
images more accessible for people with a color vision defeciency (see for 
some image samples and python code the zip archive attached).

- One algorithm does an optimized color2gray transformation trying to 
preserve more details and thus overcoming the problems of normal color2gray 
transformation

- Another algorithm does some image correction on colored images, something 
like the people on http://www.vischeck.com/daltonize/ do.

Here my questions:

1.) Is this something for the PIL standard distribution, especially the 
color2gray algorithm?

2.) My problem is: Both algorithms are very slow. You can't use them really 
on a productive system (e.g. a server where all pictures which have been 
uploaded are automattically transformed). Is there a way to speed this up? 
Is there a way to do some C-Plugin for PIL (I think I cannot optimize the 
pyhton code anymore)? I need a fast and platform independent way to 
transform images with the above algorithms (since I try to get that in the 
Moin Wiki standard distribution, see 
http://moinmo.in/FeatureRequests/EnhanceTransclusionSyntaxForImages and 
http://moinmo.in/AccessibleMoin)

I would be very happy if I could get some advice how to speed that up and/or 
how to integrate that with PIL. I am not an expert programmer. I'm at my 
wids' end.

And sorry if this mail is a littel bit "off-topic" on this mailing list. But 
I had no idea whom to ask.

Cheers,
Oliver

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ImageCorrection.zip
Type: application/x-zip-compressed
Size: 794625 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080912/f2e78459/attachment-0001.bin>

From hpj at urpla.net  Tue Sep 16 01:34:20 2008
From: hpj at urpla.net (Hans-Peter Jansen)
Date: Tue, 16 Sep 2008 01:34:20 +0200
Subject: [Image-SIG] PIL 'special' edge detection problem
Message-ID: <200809160134.21385.hpj@urpla.net>

Dear PIL-Pythonistas,

I'm trying to solve a problem with PIL which is basically related to edge 
detection, but please bear with me, since I'm pretty dump, when it comes to 
image processing apart from the very basics..

All the mess starts with a camera (Canon PowerShot 100 SX), that often 
creates erroneous images in certain twilight conditions. The ugliest failure 
is one, where parts of the image are wrapped, and color in that area is mixed 
up in some way. Hard to explain, here are some (resized) examples:
	ftp://urpla.net/cap00013.jpg
	ftp://urpla.net/cap00013-0.jpg
	ftp://urpla.net/cap00428.jpg

Ugly, isn't it. While it would be best to exchange the camera with a model, 
that doesn't suffer from these problems, that is not an option at this point.

Now, I try to detect these folds. My 'solution' misses any sophisticated 
image processing algorithm, I simply compare one pixel with the one below,
convert it to HSV (with ranges: 0..255, 0..100, 0..100), combine h and v 
deviations into a value, and try to find horizontal runs of patterns with a 
high enough value.

While the below code does it's job in some way, it's way too slow to be 
really useful, given the the original bitmap is 3264x2448 and that the camera 
is controlled by a small VIA C7 based system, which is unable to run this code 
even nearly fast enough :-(.

Does some kind soul know some better way to solve my issue (propably python
based). What I need is some special kind of horizontal edge detection, but 
where false positives are highly undesired. I know, that it should be possible 
with numpy, but maybe PIL filters may fit the job, too...

Pointers to some comprehensible theory would also be great.

Here's my code (note, that it won't work with the example images, since they
are resized, and the thresholds doesn't match anymore). The full images are
available in a subdir called fullres in the location above, as is the full 
script chkimage.py, but remember it's a test script, and it contains lots of 
unused stuff ATM..

    def chkwrappedimage(self, ipixel, width, height, fname):
        xstep = 20
        minvdev = 80
        mincnt = 12
        foldtres = 80000

        def getpixel(x, y):
            rgb = ipixel[x, y]
            hsv = RGB2HSV(*rgb)
            return rgb, hsv

        def cmppixel(hsv1, hsv2):
            h1, h2 = hsv1[0], hsv2[0]
            h = abs(h2 - h1)
            if h > 180:
                h = 360 - h
            v = abs(hsv1[2] - hsv2[2])
            return h + v

        def chkimgfold(x1, x2, xstep, y1, y2):
            foldd = {}
            for x in range(x1, x2, xstep):
                lastpix = None
                for y in range(y1, y2):
                    rgb, hsv = getpixel(x, y)
                    if lastpix != None:
                        px1, py1, rgb1, hsv1 = lastpix
                        px2, py2, rgb2, hsv2 = x, y, rgb, hsv

                        if (py1, py2) in foldd:
                            mx, lx, dv = foldd[(py1, py2)]
                            if mx >= px2:
                                lastpix = x, y, rgb, hsv
                                continue
                        else:
                            lx, dv = 0, 0
                        v = cmppixel(hsv1, hsv2)
                        if v >= minvdev:
                            while v >= minvdev:
                                px2 += 1
                                if px2 >= width:
                                    break
                                px1 = px2
                                rgb1, hsv1 = getpixel(px1, py1)
                                rgb2, hsv2 = getpixel(px2, py2)
                                v = cmppixel(hsv1, hsv2)
                                if v >= minvdev:
                                    dv += v
                            n = px2 - x
                            if n >= mincnt:
                                foldd[(py1, py2)] = px2, lx + n, dv
                    lastpix = x, y, rgb, hsv
            return foldd

        foldd = chkimgfold(0, width, xstep, 0, height)
        r = []
        for k, d in foldd.items():
            mx, lx, dv = d
            r.append((int(dv), k, lx))
        r.sort(reverse = True)

        if r:
            dv, xy, lx = r[0]
            if dv >= foldtres:
                return 'wrappedimage: %s folded at %s: value: %s (lx: %s)' % (
			fname, xy, dv, lx)

Any hints/ideas to solve this issue in a sane way are highly appreciated.

Thanks in advance,
	Pete

From G.Kloss at massey.ac.nz  Tue Sep 16 02:43:14 2008
From: G.Kloss at massey.ac.nz (Guy K. Kloss)
Date: Tue, 16 Sep 2008 12:43:14 +1200
Subject: [Image-SIG] PIL 'special' edge detection problem
In-Reply-To: <200809160134.21385.hpj@urpla.net>
References: <200809160134.21385.hpj@urpla.net>
Message-ID: <200809161243.18966.G.Kloss@massey.ac.nz>

On Tue, 16 Sep 2008 11:34:20 am Hans-Peter Jansen wrote:
> Any hints/ideas to solve this issue in a sane way are highly appreciated.

You may try to give OpenCV with its HoughLines2 implementation. Have a look 
here:

	http://opencvlibrary.sourceforge.net/CvReference#cv_imgproc_special

There are OpenCV bindings to Python, but not everything is exposed. But it may 
be worth a try as the implementation is done in native code and may work well 
for your problem.

HTH,

Guy

-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Putaiao o Mohiohio me Pangarau
Room 2.63, Quad Block A Building
Massey University, Auckland, Albany
Private Bag 102 904, North Shore Mail Centre
voice: +64 9 414-0800 ext. 9585 ? fax: +64 9 441-8181
eMail: G.Kloss at massey.ac.nz ?http://www.massey.ac.nz/~gkloss/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080916/b31247a8/attachment.pgp>

From jcupitt at gmail.com  Tue Sep 16 14:36:19 2008
From: jcupitt at gmail.com (jcupitt at gmail.com)
Date: Tue, 16 Sep 2008 13:36:19 +0100
Subject: [Image-SIG] PIL 'special' edge detection problem
In-Reply-To: <200809160134.21385.hpj@urpla.net>
References: <200809160134.21385.hpj@urpla.net>
Message-ID: <522c6460809160536yed03b3dj64b850e80ff8dcfa@mail.gmail.com>

Hi,

2008/9/16 Hans-Peter Jansen <hpj at urpla.net>:
> Now, I try to detect these folds. My 'solution' misses any sophisticated
> image processing algorithm, I simply compare one pixel with the one below,
> convert it to HSV (with ranges: 0..255, 0..100, 0..100), combine h and v
> deviations into a value, and try to find horizontal runs of patterns with a
> high enough value.

You can make this a lot quicker if you use image processing operations
rather than looping yourself for each pixel. I don't actually know the
PIL ones very well :-( but try something like:

* go to a signed monochrome image

You need to be able to handle negative values, and I don't think PIL
can do RGB int. Anyway, colour won't be much help here.

* subtract every pixel from the one below

You can do this with a convolution. Just use a mask like

  1
  -1

And each pixel will become the difference between itself and the pixel
below. Most of these values will be rather small, but if one of your
folds is present, you'll see sudden large spikes.

* take the absolute value

Now you don't need negatives!

* find the average value for each row

You'd normally do this with a projection operator (sum every row and
every column), I'm not sure if PIL has one of these. Shrinking the
image to a single column should do the same job.

* find peaks over a threshold

Now loop down the image and look for values over a certain threshold,
perhaps 50?

John

From pgretchanei at SorensonMedia.COM  Thu Sep 18 17:58:27 2008
From: pgretchanei at SorensonMedia.COM (Petro Gretchanei)
Date: Thu, 18 Sep 2008 09:58:27 -0600
Subject: [Image-SIG] PIL on intelMac
Message-ID: <1DC630CD9B5A5E428B74071D3BB210915E086F@MediaMail.MEDIA.SORENSON.COM>

Hello,
i had a question if it was possible to use PIL on mac machines? If so, where can i get the installer? 

Thank you,

Best regards,
Petro Gretchanei
Sorenson Media, Inc.
Squeeze QA, Automation Engineer

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080918/12a5f98e/attachment.htm>

From Chris.Barker at noaa.gov  Fri Sep 19 02:32:44 2008
From: Chris.Barker at noaa.gov (Christopher Barker)
Date: Thu, 18 Sep 2008 17:32:44 -0700
Subject: [Image-SIG] PIL on intelMac
In-Reply-To: <1DC630CD9B5A5E428B74071D3BB210915E086F@MediaMail.MEDIA.SORENSON.COM>
References: <1DC630CD9B5A5E428B74071D3BB210915E086F@MediaMail.MEDIA.SORENSON.COM>
Message-ID: <48D2F32C.40902@noaa.gov>

Petro Gretchanei wrote:
> Hello,
> i had a question if it was possible to use PIL on mac machines? If so, 
> where can i get the installer?

http://pythonmac.org/packages/py25-fat/index.html

note: this works with the python.org Universal Framework build, not the 
one Apple supplies (though it can be made to work with that one with a 
kludge or two -- try googling)

-Chris


> Thank you,
> 
> Best regards,
> Petro Gretchanei
> Sorenson Media, Inc.
> Squeeze QA, Automation Engineer
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig


-- 
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

From nikolai at netlife.no  Fri Sep 19 13:27:33 2008
From: nikolai at netlife.no (Nikolai Ugelvik)
Date: Fri, 19 Sep 2008 13:27:33 +0200
Subject: [Image-SIG] PIL - XVThumbImagePlugin infinite loop bugfix
Message-ID: <48D38CA5.2010006@netlife.no>

Hi.

The PIL module XVThumbImagePlugin.py has an issue with XVThumb files 
that do not include #END_OF_COMMENTS (or a comment section at all). I'm 
not quite sure if the problem is an invalid file (is the comment 
required?), but we encountered a file like this. Anyway, the result is 
XVThumbImagePlugin being stuck in an infinite loop. The patch supports 
files without comments and/or comment sections without #END_OF_COMMENTS.

Attached is a diff against XVThumbImagePlugin.py from PIL version 1.1.6. 
(I didn't figure out where else to send the diff.)

-Nikolai
-------------- next part --------------
A non-text attachment was scrubbed...
Name: XVThumbImagePlugin.diff
Type: text/x-diff
Size: 789 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080919/3f25f1a7/attachment.diff>

From cannon.el at gmail.com  Fri Sep 19 01:29:05 2008
From: cannon.el at gmail.com (Laura & Edward Cannon)
Date: Thu, 18 Sep 2008 16:29:05 -0700
Subject: [Image-SIG] PIL on intelMac
In-Reply-To: <1DC630CD9B5A5E428B74071D3BB210915E086F@MediaMail.MEDIA.SORENSON.COM>
References: <1DC630CD9B5A5E428B74071D3BB210915E086F@MediaMail.MEDIA.SORENSON.COM>
Message-ID: <b986b290809181629i728d0d0cw6dcd875d1de4d69b@mail.gmail.com>

yes it is possible to do PIL on mac.try pythonmac.org

2008/9/18 Petro Gretchanei <pgretchanei at sorensonmedia.com>

>  Hello,
> i had a question if it was possible to use PIL on mac machines? If so,
> where can i get the installer?
>
> Thank you,
>
> Best regards,
> Petro Gretchanei
> Sorenson Media, Inc.
> Squeeze QA, Automation Engineer
>
>
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080918/8464aa4a/attachment.htm>

From oliver.siemoneit at philo.uni-stuttgart.de  Fri Sep 19 23:43:12 2008
From: oliver.siemoneit at philo.uni-stuttgart.de (Oliver Siemoneit)
Date: Fri, 19 Sep 2008 23:43:12 +0200
Subject: [Image-SIG] Image Correction For Colorblind Users
Message-ID: <64475CE4B40348DC953DBC95C6411FD3@dusk>

Hi,

I have finally managed to do some C-extensions for the improved color2gray 
image transformation and the daltonize, see 
http://moinmo.in/FeatureRequests/EnhanceTransclusionSyntaxForImages for 
image samples and code.

I owe this mainly to Kevin Cazabon, which has already done a very nice 
Gaussian blur and unsharpen filter for PIL, see 
http://mail.python.org/pipermail/image-sig/2008-May/005003.html . His code 
served as a basis for my own coding work

This mail is also a reminder / request  to consider making these things part 
of a future PIL release ;-)

Oliver

----- Original Message ----- 
From: "Oliver Siemoneit" <oliver.siemoneit at philo.uni-stuttgart.de>
To: <image-sig at python.org>
Sent: Friday, September 12, 2008 6:30 PM
Subject: Image Correction For Colorblind Users


> Hi,
>
> I have implemented two image correction algorithms which probably could 
> make
> images more accessible for people with a color vision defeciency (see for
> some image samples and python code the zip archive attached).
>
> - One algorithm does an optimized color2gray transformation trying to
> preserve more details and thus overcoming the problems of normal 
> color2gray
> transformation
>
> - Another algorithm does some image correction on colored images, 
> something
> like the people on http://www.vischeck.com/daltonize/ do.
>
> Here my questions:
>
> 1.) Is this something for the PIL standard distribution, especially the
> color2gray algorithm?
>
> 2.) My problem is: Both algorithms are very slow. You can't use them 
> really
> on a productive system (e.g. a server where all pictures which have been
> uploaded are automattically transformed). Is there a way to speed this up?
> Is there a way to do some C-Plugin for PIL (I think I cannot optimize the
> pyhton code anymore)? I need a fast and platform independent way to
> transform images with the above algorithms (since I try to get that in the
> Moin Wiki standard distribution, see
> http://moinmo.in/FeatureRequests/EnhanceTransclusionSyntaxForImages and
> http://moinmo.in/AccessibleMoin)
>
> I would be very happy if I could get some advice how to speed that up 
> and/or
> how to integrate that with PIL. I am not an expert programmer. I'm at my
> wids' end.
>
> And sorry if this mail is a littel bit "off-topic" on this mailing list. 
> But
> I had no idea whom to ask.
>
> Cheers,
> Oliver
>
> 


From post at yanone.de  Wed Sep 24 13:46:37 2008
From: post at yanone.de (Yanone)
Date: Wed, 24 Sep 2008 14:46:37 +0300
Subject: [Image-SIG] Aggdraw: curveto()
Message-ID: <F952FD89-386A-4EF5-A184-6C8FA8820395@yanone.de>

hello everyone,

it seems that this is a known problem: the curveto() function of  
aggdraw draws straight lines instead of curves, ignoring the given  
bezier control points. the last post i found about this dates back to  
march 2006, and even the developer version of aggdraw via svn didn't  
change this.
if i could code in C i would have a look at it myself, but i'm helpless.

could this somehow be solved?
or is there any other way of drawing postscript curves into an image?  
the Path() method of reportlab's renderPM isn't implemented yet and i  
would like to refrain from generating a PDF file and the turn it into  
an image through imagemagick or something as i need this for a server  
implementation.

many thanks and best regards, yanone

From fredrik at pythonware.com  Wed Sep 24 20:44:57 2008
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 24 Sep 2008 20:44:57 +0200
Subject: [Image-SIG] Aggdraw: curveto()
In-Reply-To: <F952FD89-386A-4EF5-A184-6C8FA8820395@yanone.de>
References: <F952FD89-386A-4EF5-A184-6C8FA8820395@yanone.de>
Message-ID: <gbe1rb$39c$1@ger.gmane.org>

Yanone wrote:

> it seems that this is a known problem: the curveto() function of aggdraw 
> draws straight lines instead of curves, ignoring the given bezier 
> control points. the last post i found about this dates back to march 
> 2006, and even the developer version of aggdraw via svn didn't change this.
> if i could code in C i would have a look at it myself, but i'm helpless.
> 
> could this somehow be solved?

here's a workaround, at least: replace the path_close implementation 
with the following:

static PyObject*
path_close(PathObject* self, PyObject* args)
{
     if (!PyArg_ParseTuple(args, ":close"))
         return NULL;

     self->path->close_polygon();

     /* expand curves */
     agg::path_storage* path = self->path;
     agg::conv_curve<agg::path_storage> curve(*path);
     self->path = new agg::path_storage();
     self->path->add_path(curve, 0, false);
     delete path;

     Py_INCREF(Py_None);
     return Py_None;
}

(not sure this is the best way to fix this, but it should hopefully get 
you going)

</F>


From gwidion at mpc.com.br  Thu Sep 25 00:13:23 2008
From: gwidion at mpc.com.br (Joao S. O. Bueno)
Date: Wed, 24 Sep 2008 19:13:23 -0300
Subject: [Image-SIG] Aggdraw: curveto()
In-Reply-To: <F952FD89-386A-4EF5-A184-6C8FA8820395@yanone.de>
References: <F952FD89-386A-4EF5-A184-6C8FA8820395@yanone.de>
Message-ID: <200809241913.23786.gwidion@mpc.com.br>

On Wednesday 24 September 2008, Yanone wrote:
> hello everyone,
>
> it seems that this is a known problem: the curveto() function of
> aggdraw draws straight lines instead of curves, ignoring the given
> bezier control points. the last post i found about this dates back to
> march 2006, and even the developer version of aggdraw via svn didn't
> change this.
> if i could code in C i would have a look at it myself, but i'm helpless.
>
> could this somehow be solved?
> or is there any other way of drawing postscript curves into an image?
> the Path() method of reportlab's renderPM isn't implemented yet and i
> would like to refrain from generating a PDF file and the turn it into
> an image through imagemagick or something as i need this for a server
> implementation.
>
> many thanks and best regards, yanone

Hi there - 
Sorry,  i missed some of the context for this thread, so I don't? know what 
you have already tried -
but you might try using the Cairo Library (and its python bingins, of course)
It does implement postscript drawing operators and can render to many 
back-ends.

	js
	-><-




From post at yanone.de  Thu Sep 25 00:32:30 2008
From: post at yanone.de (Yanone)
Date: Thu, 25 Sep 2008 01:32:30 +0300
Subject: [Image-SIG] Aggdraw: curveto()
Message-ID: <93BD5F6E-17AC-4AD3-AB05-32328F5A37C8@yanone.de>

thanks a lot, it worked. it draws the outlines upside down, which i  
guess is just due to a different corner of (0,0) of the two canvases  
(vector and PIL). in case this is the case, i'll work around it by  
simply turning all y coordinates.
many thanks, yanone

From starsareblueandfaraway at gmail.com  Thu Sep 25 23:53:30 2008
From: starsareblueandfaraway at gmail.com (Roy H. Han)
Date: Thu, 25 Sep 2008 17:53:30 -0400
Subject: [Image-SIG] How do I load and display a 16 bit grayscale image?
Message-ID: <6a5569ec0809251453u3fd7930yfe6274c8b163f092@mail.gmail.com>

Hi,

I'm having some trouble displaying an image and I was wondering if someone
here can help.

Specifically, I loaded a 16 bit grayscale image using the fromstring()
method with mode I;16.  When I try to display the image, the entire image is
white with some specks of black.  I think it is because PIL thinks that it
is a color image, but I'm not sure.

Is there a way to tell PIL that the 16 bit image is actually grayscale?

The details are that I am opening GeoTIFFs using GDAL and using PIL's
fromstring() to see the resulting images.  The multichromatic image seems to
display fine when I view each band individually (PIL's merge doesn't seem to
work for 16 bit images yet).  However, the panchromatic image is washed out
in white and I can't see anything.  When I open the panchromatic image in
qGIS, however, all the trees are plainly visible.

Thanks,
Roy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080925/d6451d94/attachment.htm>

From szymon at mwg.pl  Mon Sep 29 01:30:26 2008
From: szymon at mwg.pl (Szymon Kosok)
Date: Mon, 29 Sep 2008 01:30:26 +0200
Subject: [Image-SIG] No "transparency" key in info for transparent PNG
Message-ID: <199b51b30809281630o1abf9b56ubef76dd4c3a9720c@mail.gmail.com>

Hello,

I have transparent PNG (when I open it in Photoshop it's transparent)
and when I try to open it with PIL there is no "transparency" key in
info. I've tried another transparent PNG (famfamfam icon) and no luck
with that too. What can cause that problem? Mode is "RGBA".

Best regards,
Szymon

From cannon.el at gmail.com  Mon Sep 29 04:07:47 2008
From: cannon.el at gmail.com (Laura & Edward Cannon)
Date: Sun, 28 Sep 2008 19:07:47 -0700
Subject: [Image-SIG] No "transparency" key in info for transparent PNG
In-Reply-To: <199b51b30809281630o1abf9b56ubef76dd4c3a9720c@mail.gmail.com>
References: <199b51b30809281630o1abf9b56ubef76dd4c3a9720c@mail.gmail.com>
Message-ID: <b986b290809281907n368adc16m7a21342b9bbe1812@mail.gmail.com>

As far as I can tell, you should be able to determine whether or not
the image has transparency by looking at the mode. if it is RGBA or LA
it has transparency, otherwise it does not. If you need to know what
parts are transparent, look at the mask, obtained by im.split()

On Sun, Sep 28, 2008 at 4:30 PM, Szymon Kosok <szymon at mwg.pl> wrote:
> Hello,
>
> I have transparent PNG (when I open it in Photoshop it's transparent)
> and when I try to open it with PIL there is no "transparency" key in
> info. I've tried another transparent PNG (famfamfam icon) and no luck
> with that too. What can cause that problem? Mode is "RGBA".
>
> Best regards,
> Szymon
> _______________________________________________
> Image-SIG maillist  -  Image-SIG at python.org
> http://mail.python.org/mailman/listinfo/image-sig
>

From ustrum at gmail.com  Mon Sep 29 09:42:36 2008
From: ustrum at gmail.com (Pablo Martin Rodriguez)
Date: Mon, 29 Sep 2008 09:42:36 +0200
Subject: [Image-SIG] Splines
Message-ID: <a6dac2250809290042v2d40333cg7d9c3a84fa93b68e@mail.gmail.com>

Hi all

I need to draw splines and I don't know how to do it using PIL. The only
idea that I have had is to calculate all the points in the spline, given by
the user a couple of points, and then draw them with the ImageDraw.point,
but it doesn't seem to be a very good option. Any idea?? Maybe using another
library is better than using PIL?? I'm sure that somebody has had to draw
splines using python before, but I've found nothing when searching in
google. Thanks a million!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/image-sig/attachments/20080929/1acaaf30/attachment.htm>

From Chris.Barker at noaa.gov  Tue Sep 30 06:50:51 2008
From: Chris.Barker at noaa.gov (Christopher Barker)
Date: Mon, 29 Sep 2008 21:50:51 -0700
Subject: [Image-SIG] Splines
In-Reply-To: <a6dac2250809290042v2d40333cg7d9c3a84fa93b68e@mail.gmail.com>
References: <a6dac2250809290042v2d40333cg7d9c3a84fa93b68e@mail.gmail.com>
Message-ID: <48E1B02B.30700@noaa.gov>

Pablo Martin Rodriguez wrote:
> I need to draw splines and I don't know how to do it using PIL. 

http://effbot.org/zone/pythondoc-aggdraw.htm

You want to use a Path object, and the curveto() method.

-Chris


-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (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