Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.python > #71780 > unrolled thread

Clip Raster Image Pair by Overlapping Area using Python

Started bymikejohnryan08@gmail.com
First post2014-05-19 18:05 -0700
Last post2014-05-19 21:22 -0700
Articles 5 — 3 participants

Back to article view | Back to comp.lang.python


Contents

  Clip Raster Image Pair by Overlapping Area using Python mikejohnryan08@gmail.com - 2014-05-19 18:05 -0700
    Re: Clip Raster Image Pair by Overlapping Area using Python Chris Angelico <rosuav@gmail.com> - 2014-05-20 11:17 +1000
      Re: Clip Raster Image Pair by Overlapping Area using Python mikejohnryan08@gmail.com - 2014-05-19 18:30 -0700
        Re: Clip Raster Image Pair by Overlapping Area using Python Chris Angelico <rosuav@gmail.com> - 2014-05-20 11:43 +1000
          Re: Clip Raster Image Pair by Overlapping Area using Python Rustom Mody <rustompmody@gmail.com> - 2014-05-19 21:22 -0700

#71780 — Clip Raster Image Pair by Overlapping Area using Python

Frommikejohnryan08@gmail.com
Date2014-05-19 18:05 -0700
SubjectClip Raster Image Pair by Overlapping Area using Python
Message-ID<96220021-03b7-4016-bbf1-f660b9590dab@googlegroups.com>
Hello,

I have a set of aerial images which I am trying to clip by their overlapping areas, for use in a more involved program I am writing with PIL.

What would be the best method with Python to extract the overlapping area from a pair of images?  I know there are libraries out there that could do this, but I couldn't find any so far.

I need to execute this through Python as it will be part of a batch processing script.  Any ideas/advice would be great!  Thanks so much in advance. 

[toc] | [next] | [standalone]


#71781

FromChris Angelico <rosuav@gmail.com>
Date2014-05-20 11:17 +1000
Message-ID<mailman.10145.1400548633.18130.python-list@python.org>
In reply to#71780
On Tue, May 20, 2014 at 11:05 AM,  <mikejohnryan08@gmail.com> wrote:
> I have a set of aerial images which I am trying to clip by their overlapping areas, for use in a more involved program I am writing with PIL.
>
> What would be the best method with Python to extract the overlapping area from a pair of images?  I know there are libraries out there that could do this, but I couldn't find any so far.
>

The problem is that they won't absolutely perfectly match, so you need
some sort of similarity check. That's hard - very hard. I suggest you
find a stand-alone program that will do that part of the job; it'll be
easier than finding a Python library for it.

ChrisA

[toc] | [prev] | [next] | [standalone]


#71782

Frommikejohnryan08@gmail.com
Date2014-05-19 18:30 -0700
Message-ID<c28d9a2d-ea4a-4878-927b-4855fdc5ce70@googlegroups.com>
In reply to#71781
On Monday, May 19, 2014 9:17:05 PM UTC-4, Chris Angelico wrote:
> On Tue, May 20, 2014 at 11:05 AM,  <mikejohnryan08@gmail.com> wrote:
> 
> > I have a set of aerial images which I am trying to clip by their overlapping areas, for use in a more involved program I am writing with PIL.
> 
> >
> 
> > What would be the best method with Python to extract the overlapping area from a pair of images?  I know there are libraries out there that could do this, but I couldn't find any so far.
> 
> >
> 
> 
> 
> The problem is that they won't absolutely perfectly match, so you need
> 
> some sort of similarity check. That's hard - very hard. I suggest you
> 
> find a stand-alone program that will do that part of the job; it'll be
> 
> easier than finding a Python library for it.
> 
> 
> 
> ChrisA


Chris,

Thanks for your response.  For my purpose, the images won't have to be 'perfectly' matched, but hopefully as close as possible.  Registration algorithms won't work for this reason--the images aren't identical to each other because of the different perspectives.

[toc] | [prev] | [next] | [standalone]


#71783

FromChris Angelico <rosuav@gmail.com>
Date2014-05-20 11:43 +1000
Message-ID<mailman.10146.1400550231.18130.python-list@python.org>
In reply to#71782
On Tue, May 20, 2014 at 11:30 AM,  <mikejohnryan08@gmail.com> wrote:
> Thanks for your response.  For my purpose, the images won't have to be 'perfectly' matched, but hopefully as close as possible.  Registration algorithms won't work for this reason--the images aren't identical to each other because of the different perspectives.
>

Yeah, and that's what makes it hard. This is outside my area of
expertise, so all I can say is that these sorts of programs do exist,
but restricting yourself to Python libraries may make your life
unnecessarily hard.

On a separate topic, you're posting through Google Groups with its
abhorrent bugs. Can you please either edit your posts before sending
(removing the blank lines, wrapping to a sane width, and trimming the
quoted text), or switch to a better newsreader (or the mailing list)?
You'll find that a lot of people here either filter your message to
/dev/null automatically, or take one look and just move on to the next
post. Remember, never offend people with form when you can offend them
with substance :)

ChrisA

[toc] | [prev] | [next] | [standalone]


#71787

FromRustom Mody <rustompmody@gmail.com>
Date2014-05-19 21:22 -0700
Message-ID<f9a71862-ba3d-44bf-8dfe-45b1a6241011@googlegroups.com>
In reply to#71783
On Tuesday, May 20, 2014 7:13:42 AM UTC+5:30, Chris Angelico wrote:

> On a separate topic, you're posting through Google Groups with its
> abhorrent bugs. Can you please either edit your posts before sending
> (removing the blank lines, wrapping to a sane width, and trimming the
> quoted text), or switch to a better newsreader (or the mailing list)?
> You'll find that a lot of people here either filter your message to
> /dev/null automatically, or take one look and just move on to the next
> post. Remember, never offend people with form when you can offend them
> with substance :)

This https://wiki.python.org/moin/GoogleGroupsPython
shows the problems and possible solutions

[toc] | [prev] | [standalone]


Back to top | Article view | comp.lang.python


csiph-web