Moiré artefact reduction in Talbot-Lau X-ray imaging

Abstract

Talbot-Lau X-ray imaging provides information about X-ray scattering, refraction, and attenuation features of objects. This capability offers a variety of possibilities for medical and industrial applications. However, already slight inaccuracies during the measurement process result in moiré artefacts in the reconstructed images. A sufficient setup stability is expensive and not invariably achievable. We developed an advanced reconstruction algorithm, which reduces moiré artefacts by compensating these inaccuracies. Therefore, we defined a cost function, which is sensitive to moiré artefacts. This cost function is minimized by adjusting the phase-step positions. To demonstrate the capability of the developed algorithm, we executed an acquisition of a pig's trotter while the setup was deliberately disturbed by vibrations. By applying the developed algorithm, the moiré artefacts were largely reduced. This development is a crucial step to facilitate Talbot-Lau X-ray imaging in clinical practice and industrial applications.