search

FranckLab / FIDVC

Fast Iterative Digital Volume Correlation Algorithm
31 stars 18 forks source link

Registration 3D images using FIDVC #27

Closed schweini2409 closed 4 years ago

schweini2409 commented 4 years ago

Hello, I'd like to align 2 stacks of images so I'm wondering whether I can use FIDVC to find the displacement between the two stacks then apply this displacement to one of the images for the registration? Thank you,

mohakpatel commented 4 years ago

Yes, that should work. You will have to take mode of the displacement to get the most accurate estimation of the rigid displacement between the two images.

If you have 2D images, you can use https://github.com/FranckLab/qDIC.

schweini2409 commented 4 years ago

Then I suppose that I can use the volumeMapping.m to apply the displacement found previously to the volume ?

ALandauer commented 4 years ago

The volumeMapping function is probably a bit too specialized since it does both the forward and backward mapping. A simpler solution is to either re-interpolate onto a shifted grid or use the Matlab built-in affine3d (and related) function.

For an example of re-interpolation look at volumeMapping, but only implement the "forward" part with a single uniform value of displacement measured from running FIDVC. Although, you can use Matlab built-in interpolation instead of the mex interpolation, since you'll only need to do it once.

On Tue, Jul 14, 2020 at 4:39 AM Tuan Tu NGUYEN notifications@github.com wrote:

Then I suppose that I can use the volumeMapping.m to apply the displacement found previously to the volume ?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/FranckLab/FIDVC/issues/27#issuecomment-658052203, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADCNG3B4LH3DYEXH4FCYPYDR3QKTPANCNFSM4OV6OUYA .