Iterative reconstruction in x-ray computed laminography from differential phase measurements

Area of Science:

• X-ray imaging and computed tomography
• Materials science and engineering
• Diffractive optics

Background:

• Traditional X-ray computed tomography is unsuitable for extended flat objects.
• Phase-contrast imaging offers enhanced sensitivity for materials.
• Laminography presents unique challenges due to its specific geometry.

Purpose of the Study:

• To demonstrate phase-contrast X-ray computed laminography for extended flat objects.
• To develop and evaluate an iterative reconstruction algorithm for improved spatial resolution.
• To compare iterative reconstruction with filtered backprojection for laminographic data.

Main Methods:

• Utilized a Talbot interferometer for differential phase measurements.
• Derived filtered backprojection formula for differential phase data.
• Implemented an iterative filtered backprojection algorithm incorporating prior support information.
• Employed GPU hardware for efficient volume reconstruction.
• Conducted synchrotron radiation experiments on polymer meshes.

Main Results:

• Successfully reconstructed the complex refractive index of extended flat objects.
• Demonstrated improved spatial resolution in the direction normal to the object plane using the iterative algorithm.
• Quantified the degradation of spatial resolution inherent in laminography's reciprocal space sampling.
• Showcased the efficacy of iterative reconstruction over simpler filtered backprojection for polymer mesh samples.

Conclusions:

• Phase-contrast X-ray computed laminography is a viable technique for extended flat objects.
• Iterative reconstruction algorithms significantly enhance spatial resolution in laminography.
• GPU acceleration is crucial for practical implementation of these advanced reconstruction methods.