images to evaluate segmentation algorithm

Thread View: sci.image.processing

2 messages

[T] Back to Threads [V] Tree View [G] Group Articles

2 total messages Started by Patrick Moran Tue, 06 Oct 1992 03:11

#660

Author: Patrick Moran
Date: Tue, 06 Oct 1992 03:11

24 lines
1762 bytes

Hello,

I'm looking for a few images to use for some experiments with a new
image segmentation approach.  Currently I'd like to find a few
remote sensing images with intensities at two or three wavelengths.
I've already found gif files of boston, new york and so on (for
example from vab02.larc.nasa.gov), but they aren't quite what I'm looking
for.  What I'd really like are a few simpler images where there are
larger, roughly homogeneous regions such as "corn," "soybeans," "water,"
"roadway," and so on.  The images I've found so far have easily
discernable "water" regions but the rest is much harder.  Best of
all for me would be images that are accompanied by a "correct"
segmentation, i.e. a standard to measure my results against.

I realize that it isn't too likely that what I'm looking for is
publically available, but I'd appreciate any help that could be
provided.

Pat

-- 
Patrick Moran  (pmoran@ncsa.uiuc.edu)                 4059 Beckman Institute
NCSA Biomedical Imaging Group                         405 N. Mathews Avenue
University of Illinois at Urbana-Champaign            Urbana, Illinois 61801

Re: images to evaluate segmentation algorithm

#670

Author: Hans DU BUF
Date: Tue, 06 Oct 1992 13:35

34 lines
1987 bytes

In article <BvoJJr.EIA@news.cso.uiuc.edu>, pmoran@talon.ncsa.uiuc.edu (Patrick Moran) writes:
|> Hello,
|> 
|> I'm looking for a few images to use for some experiments with a new
|> image segmentation approach.  Currently I'd like to find a few
|> remote sensing images with intensities at two or three wavelengths.
|> I've already found gif files of boston, new york and so on (for
|> example from vab02.larc.nasa.gov), but they aren't quite what I'm looking
|> for.  What I'd really like are a few simpler images where there are
|> larger, roughly homogeneous regions such as "corn," "soybeans," "water,"
|> "roadway," and so on.  The images I've found so far have easily
|> discernable "water" regions but the rest is much harder.  Best of
|> all for me would be images that are accompanied by a "correct"
|> segmentation, i.e. a standard to measure my results against.
|> 
|> I realize that it isn't too likely that what I'm looking for is
|> publically available, but I'd appreciate any help that could be
|> provided.
|> 
|> Pat
|> 
|> -- 
|> Patrick Moran  (pmoran@ncsa.uiuc.edu)                 4059 Beckman Institute
|> NCSA Biomedical Imaging Group                         405 N. Mathews Avenue
|> University of Illinois at Urbana-Champaign            Urbana, Illinois 61801

Our test images with aerial (NOT satellite) textures can be found at

pioneer.unm.edu (129.24.9.217) with anonym ftp
directory: pub/texture_maps
see also the README file there

regards,
Hans

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads