On super-resolution holography: effective geometry, sampling, and constraints

X-ray holo-tomography is a powerful tool for high-resolution three-dimensional imaging, but it faces inherent limitations in resolution caused by detector constraints, numerical aperture restrictions, and the empty beam division, which is a ubiquitous step of raw data correction in conventional inline holography and (phase contrast) <i>μ</i>CT. To address these challenges, we extend the technique of super-resolution holography (SRH), an iterative phase reconstruction scheme that achieves high resolution by leveraging a complex illumination model, compatible with low flux areas in the beam tail and far-field diffraction components. We include a realistic pixel block constraint and represent the fields with moderate upsampling in the effective parallel beam geometry. We then demonstrate the improvements gained in resolution and sampling efficiency experimentally, using X-ray waveguide illumination and a single photon counting detector. Our results highlight the benefits of combining computational development with experimental design and showcase the potential of SRH to improve resolution and contrast at holo-tomography beamlines.