Abstract
X-Ray Computed Tomography (XCT) techniques have evolved to a point that high-resolution data can be acquired so fast that classic segmentation methods are prohibitively cumbersome, demanding automated data pipelines capable of dealing with non-trivial 3D images. Meanwhile, deep learning has demonstrated success in many image processing tasks, including materials science applications, showing a promising alternative for a human-free segmentation pipeline. However, the rapidly increasing number of available architectures can be a serious drag to the wide adoption of this type of models by the end user. In this paper a modular interpretation of U-Net (Modular U-Net) is proposed with a parametrized architecture that can be easily tuned to optimize it. As an example, the model is trained to segment 3D tomography images of a three-phased glass fiber-reinforced Polyamide 66. We compare 2D and 3D versions of our model, finding that the former is slightly better than the latter. We observe that human-comparable results can be achievied even with only 13 annotated slices and using a shallow U-Net yields better results than a deeper one. As a consequence, neural networks show indeed a promising venue to automate XCT data processing pipelines needing no human, adhoc intervention.
Highlights
X-ray Computed Tomography (XCT), a characterization technique used by material scientists for non-invasive analysis, has tremendously progressed over the last 10 years with improvements in both spatial resolution and throughput (Withers et al, 2021; Maire and Withers, 2014)
The data used in this work is composed of synchrotron X-ray tomography volumes recorded using 2 mm × 2 mm cross section composite specimens of Polyamide 66 reinforced by glass fibers, a material commonly used for structural pieces in different applications
In terms of processing speed, the results show that our method could be carried out almost in real time using typical hardware available at a synchrotron beamline: using an NVIDIA Quadro P2000 (5 GB), it took 32 min to process the volume Crack, with approximately 5.8 billion voxels (1579 × 1845 × 2002)
Summary
X-ray Computed Tomography (XCT), a characterization technique used by material scientists for non-invasive analysis, has tremendously progressed over the last 10 years with improvements in both spatial resolution and throughput (Withers et al, 2021; Maire and Withers, 2014). Progress with synchrotron sources, including the recent European Synchrotron Radiation Facility (ESRF) upgrade (Pacchioni, 2019), made it possible to look inside a specimen without destroying it in a matter of seconds (Shuai et al, 2016)—sometimes even faster (Maire et al, 2016). This results in a wealth of 3D tomography images (stack of 2D images) that need to be analyzed and, in some applications, it is desirable to segment them (i.e., transform the gray-scaled voxels into semantic categorical values).
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.