Automatic extraction of dynamic features from sub-second tomographic microscopy data

A fully automatized iterative reconstruction pipeline (rSIRT-PWC-DIFF) was developed at TOMCAT to reconstruct and segment dynamic features from a static matrix. The proposed algorithm protocol separates dynamic features automatically through difference sinograms. The implemented virtual sinogram step enables direct applicability to interior tomography datasets while time-regularization is performed on small sub-regions to increase algorithm robustness on severely limited signal-to-noise ratio datasets. The automatic stopping criterion guarantees full independence from manual parameter tuning. The advantages of the proposed algorithm are demonstrated on dynamic fuel cell data, for which the current data post-processing pipeline heavily relies on manual labor. The proposed approach reduces the post-processing time by at least a factor of 4 on limited computational resources. Full independence from manual interaction additionally allows straightforward up-scaling to efficiently process larger data, extensively boosting the possibilities in future dynamic X-ray tomographic investigations.