X-ray computed tomography for materials characterization: Transformation by machine learning

A. Takase
Rigaku Corporation,
United States

Keywords: X-ray, computed tomography, materials characterization, machine learning


It has been almost 100 years since Radon proved that X-ray computed tomography (CT) worked in concept. Much progress has been made since then. And one of the most significant progress was made in the last several years. Even after X-ray CT image quality improved by better X-ray generators and detectors, image segmentation, the very first step of quantitative analysis, remained a challenge for decades. However, image segmentation by machine learning, including deep learning, has proved to handle some of the most challenging image segmentation tasks. Today, this technique is not only for CT specialists but is available as a tool for anyone to use. This talk will briefly cover the basics of X-ray CT and discuss how machine learning segmentation provides new possibilities for X-ray CT applications.