When images are coregistered, there should not normally be any
isotropic scaling introduced. I'm therefore wondering which matrix
you are trying to adjust.
In general, when trying to factor out zooms and shears from an affine
transform matrix, you would use some form of ordinary Procrustes
analysis ( http://en.wikipedia.org/wiki/Procrustes_analysis ). This
needs additional information about which regions of the coordinate
system should change the least. The usual descriptions of Procrustes
analysis refer to landmarks, but it is also possible to use some form
of mask image (of the brain?) to achieve similar results.
Best regards,
-John
On 16 June 2011 01:07, Aaryani Tipirneni <[log in to unmask]> wrote:
> Hi all
>
> Could you please tell me how to remove the zooming-in/zooming-out factor (scaling) from the affine transformation matrix? Though the scaling is removed, I want the coregistration results to be same.
>
> I tried P = spm_imatrix(M) ; M- affine transformation matrix
> P(7:9)=1;
>
> But it did not work! Please tell me if I need to change something else.
>
>
> Please help!
>
> Thanks
> Aaryani
>
|