comparing 2 images

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

comparing 2 images

Holly Treddenick
Hi
I am using Image J to help analyze static posture. I took a picture of 30
people standing neutral with some markers on bony landmarks on their body.
Half of them were put through a 4 week treatment period and half of them
weren't. They all came back and got photographed again, in the same neutral
stance. I know Image J can be used to compare the 2 images to measure the
changes, but I haven't been able to figure out how. I tried writing down the
xy co-ordinates, but I'm sure there is a better way.

Also, the angle of the camera was changed for a few of the pictures, but the
distance always remained the same. Can anyone help guide me in this process?
Thank you!

Thanks so much!
Holly
[hidden email]

--
Holly Treddenick
Blue Bird Osteopathy
Femmes du Feu
26 Sarnia Ave
647-234-2583
Reply | Threaded
Open this post in threaded view
|

Re: comparing 2 images

Albert Cardona
> Also, the angle of the camera was changed for a few of the pictures, but
> the
> distance always remained the same. Can anyone help guide me in this
> process?


You need to scale the images relative to the common landmarks. Stephan
Saalfeld will demonstrate a plugin on how to do this automatically at the
ImageJ conference this November 6-7 in Luxemburg.

It has to do with feature extraction:

  http://pacific.mpi-cbg.de/wiki/index.php/Feature_Extraction

... and in extracting the transformation matrix from the feature
correspondences, and then registering both images:

  http://pacific.mpi-cbg.de/wiki/index.php/Plugins_Menu#Registration

Hope that helped a bit.

Albert
--
Albert Cardona
http://www.mcdb.ucla.edu/Research/Hartenstein/acardona/
Reply | Threaded
Open this post in threaded view
|

Re: comparing 2 images

karo03
In reply to this post by Holly Treddenick
Possibly your aproach is better handeled by a landmark-based analysis.
See for that under http://www.morphometrics.org/

Regards
Karsten

Am 16.10.2008 um 22:35 schrieb Holly Treddenick:

> Hi
> I am using Image J to help analyze static posture. I took a picture  
> of 30
> people standing neutral with some markers on bony landmarks on their  
> body.
> Half of them were put through a 4 week treatment period and half of  
> them
> weren't. They all came back and got photographed again, in the same  
> neutral
> stance. I know Image J can be used to compare the 2 images to  
> measure the
> changes, but I haven't been able to figure out how. I tried writing  
> down the
> xy co-ordinates, but I'm sure there is a better way.
>
> Also, the angle of the camera was changed for a few of the pictures,  
> but the
> distance always remained the same. Can anyone help guide me in this  
> process?
> Thank you!
>
> Thanks so much!
> Holly
> [hidden email]
>
> --
> Holly Treddenick
> Blue Bird Osteopathy
> Femmes du Feu
> 26 Sarnia Ave
> 647-234-2583
Reply | Threaded
Open this post in threaded view
|

Re: comparing 2 images

Marcel
<quote author="Karsten Rodenacker-3">
"Possibly your aproach is better handeled by a landmark-based analysis.
See for that under http://www.morphometrics.org/


Regards
Karsten"


Apropos morphometrics.
Recently a good book (im my opinion) was published about morphometrics with R
which can be used in combination with ImageJ (the book examples not using Imagej!).
http://www.springer.com/statistics/stats+life+sci/book/978-0-387-77789-4
Reply | Threaded
Open this post in threaded view
|

Re: comparing 2 images

Jeff Linn
In reply to this post by Holly Treddenick
I've been doing this for years in my Rolfing practice and class, first
with NIH Image and lately with ImageJ.  I've developed a set of macros
that work with the Mac and a firewire video camera or with Linux and a
digital camera (downloading the pictures.  I'd be happy to share the
macros I have (they are a little rough but available on my website at
http://jefflinn.yogabodywork.net/imaging.html ).

As far as adjusting for different camera positions in the pictures there
are better minds on this board for that. I learned early on that it was
much easier to have the camera in a fixed position rather than post
process.  I think there was an article written in the Journal of
Bodywork and Movement Therapies a while back (not sure when but you may
be able to search them online) about doing this with ImageJ (technical
no macros or plugins as I remember).

Good Luck,

Jeff Linn
Guild for Structural Integration
[hidden email]

On Thu, 2008-10-16 at 16:35 -0400, Holly Treddenick wrote:

> Hi
> I am using Image J to help analyze static posture. I took a picture of 30
> people standing neutral with some markers on bony landmarks on their body.
> Half of them were put through a 4 week treatment period and half of them
> weren't. They all came back and got photographed again, in the same neutral
> stance. I know Image J can be used to compare the 2 images to measure the
> changes, but I haven't been able to figure out how. I tried writing down the
> xy co-ordinates, but I'm sure there is a better way.
>
> Also, the angle of the camera was changed for a few of the pictures, but the
> distance always remained the same. Can anyone help guide me in this process?
> Thank you!
>
> Thanks so much!
> Holly
> [hidden email]
>