Padding without extending image size to increase the field of view

I’m trying to add an uniform margin to a 3D CT volume using itk::ConstantPadImageFilter and then resize into original image size.
My ultimate goal is ,for example, to transform the bottom image to the image at the top to increase the field of view without changing the original image size.

This can be accomplished using resample filter. Setting the original image as the reference should work well for your case. But you will need a transform with some scaling, in order to have non-trivial effect.


No. I don’t have two images. I only have set of images like in the bottom image and I need to transform a way that it appears with a large field of view. I tried by adding extra black pixels to the original image but the original image size become increasing in that case. I then need to pass this 3D volume to the SiddonJacobRay tracing algorithm to generate DRR images by capturing the whole chest. Any solution for that ?

1 Like

Any solution for this please ?


It’s still sounds like @dzenanz answer is still appropriate. Perhaps reviewing some of ITK Image, Transform and resampling concepts will help:

1 Like