Overcoming the data processing challenges of unifying tomography techniques at ESRF

Abstract

The ESRF tomography software is undergoing a major re-write, targeted at unifying the tomography user experience, data acquisition, data format, and processing tools. To cope with the inherent large complexity of handling data coming from multiple beamlines, techniques and facilities, we are developing an open source software called NXtomomill. It is named after the NXtomo application class of the NeXus data format, and it will offer advanced data conversion, manipulation and reduction functions. This includes azimuthal integration for X-ray Diffraction CT, and elemental fitting for X-ray Fluorescence CT.This work will have a remarkable impact on resource management, software robustness, and data storage. With the addition of the new functionality of the ESRF data portal, it will also unlock unimagined opportunities with respect to the automation of artificial intelligence techniques on large and heterogeneous collections of datasets.