Alignment solutions in CT image processing
Virtual Alignment Method (VAM) for projection image alignment algorithm to get clean CT image reconstructions.
- VAM for rigid sample
- VAM for elastic sample
- Generate sinogram using Radon Transform for Shepp-Logan phantom image
- Apply random translation error to each column of sinogram
- Calculate center of attenuation and use it as a fixed point
- Apply VAM to the obtained fixed point
- CT image reconstruction acquisition
P.S. See "fp_phantom.png" for the location of the fixed point in the reconstruction
- Calculate a fixed point in the whole projection set
- Align the projection image set to the axial level (common layer) of the specimen by adjusting the height of the fixed point
- Apply VAM to a fixed point on the adjusted projection image set
- Obtain ideally aligned CT reconstructions using an ideally aligned projection image set