Fast X-ray micro-tomography for low-Z materials

Abstract

X-ray phase-contrast micro-tomography (XPCMT) is an important method for the non-destructive acquisition of internal information from samples composed of low-Z elements. During the development of XPCMT, its spatial resolution has gradually been improved; however, insufficient attention has been directed towards the improvement of its time resolution. The low time resolution of XPCMT restricts its applications in fast dynamic processes, such as the fermentation process and alloy growth. In this paper, we demonstrate a fast XPCMT method developed by combining the compressed sensing (CS) theory with XPCMT. This method allows for the accurate reconstruction of images using undersampled XPCMT data, thus achieving higher time resolution and simultaneously reducing the dose delivered to the samples; the latter is especially beneficial in medical applications. The CS-XPCMT algorithm was validated using experimental data from two samples, Fructus Foeniculi and a live ant, collected at the X-ray imaging and biomedical application beamline of the Shanghai Synchrotron Radiation Facility. The results for Fructus Foeniculi demonstrate that the CS-XPCMT algorithm yields good reconstruction accuracy for incomplete and undersampled data. Furthermore, the results for the live ant demonstrate that the CS-XPCMT algorithm is capable of performing fast XPCMT and is a potential method for the realisation of dynamic XPCMT, given appropriately upgraded experimental devices.