Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

High-speed X-ray imaging and 3D analysis of impact-formed fragments

 
: Moser, Stefan; Nau, Siegfried; Wickert, Matthias

:

Versluis, M. ; Society of Photo-Optical Instrumentation Engineers -SPIE-, Bellingham/Wash.:
32nd International Congress on High-Speed Imaging and Photonics 2018 : 8-12 October 2018, Enschede, The Netherlands
Bellingham, WA: SPIE, 2019 (Proceedings of SPIE 11051)
ISBN: 978-1-5106-2777-2
ISBN: 978-1-5106-2778-9
Paper 110510O, 7 S.
International Congress on High-Speed Imaging and Photonics <32, 2018, Enschede>
Englisch
Konferenzbeitrag
Fraunhofer EMI ()

Abstract
For high-speed impact incidents associated with the generation of flashes, dust or smoke, optical imaging often is not feasible. In these cases, X-ray flash imaging is one of the established tools for the analysis. However, Xray flash imaging is often limited to few (usually max. 8) images from one or equally few directions. This severely limits the spatial resolution and capacity to derive 3D information about the observed process as well as temporal resolution to capture the dynamics of the process. Usually, several X-ray images are taken at one point in time from different perspectives to create a 3D-reconstruction from few projections (High-Speed X-ray Computed Tomography, HSCT), or taken at several points in time from the same perspective to allow the evaluation of the dynamics. For some applications, this is not sufficient. In this talk, we demonstrate how these limitations can be overcome, by presenting an exemplary approach to investigate impact processes with X-ray flash imaging. By the use of a priori information about the process, it is possible to gain the flight trajectory including dynamic information (velocity) for each observed fragment from one single experiment. This is possible by acquiring several X-ray flash images from different viewing directions at different points in time. By using a data fusion scoring algorithm, possible dynamic solutions for each fragment are found and can be evaluated to characterize the process. Using this method, fragments on the size of < 1 cm can be localized to trajectories with a statistical error of ± 5 mm and a statistical relative velocity error of < 5%. In an outlook, we evaluate how this approach can be transferred to other measurement methods.

: http://publica.fraunhofer.de/dokumente/N-534684.html