MSSP - Point Cloud.pdf (3.09 MB)
Download filePoint cloud-based elastic reverse time migration for ultrasonic imaging of components with vertical surfaces
journal contribution
posted on 2021-06-22, 09:23 authored by Jing Rao, Jilai Wang, Stefan Kollmannsberger, Jianfeng Shi, Hailing Fu, Ernst RankThis work presents a new ultrasonic imaging framework for non-destructive evaluation of components with vertical or steeply dipping surfaces and demonstrates its ability of accurately characterizing multiple defects hidden in the interior of the component based on a limited coverage of ultrasonic linear phased array. Central to the framework is a point cloud-based elastic reverse time migration (PC-based ERTM) method. First, a surface reconstruction is derived from the point cloud provided through photos of an object from multiple views by bundle adjustment. Second, by taking the surface reconstruction as a geometric background estimate for elastic reverse time migration, the algorithm considers information of multiple scattering and mode conversions as well as multiple wave reflections from the component’s bottom and aims at detecting internal defects. The effectiveness and accuracy of the PC-based ERTM approach is examined by experiments with multiple defects in extruded aluminum specimens with vertical surfaces. Experimental results show that notches and side-drilled holes in components can be reconstructed accurately.
Funding
Alexander von Humboldt Foundation under grant number 1022809
German Research Foundation (DFG) under the Grant No. Ra624/29-1 and KO4570/1-1
History
School
- Mechanical, Electrical and Manufacturing Engineering
Published in
Mechanical Systems and Signal ProcessingVolume
163Publisher
Elsevier BVVersion
- AM (Accepted Manuscript)
Rights holder
© ElsevierPublisher statement
This paper was accepted for publication in the journal Mechanical Systems and Signal Processing and the definitive published version is available at https://doi.org/10.1016/j.ymssp.2021.108144.Acceptance date
2021-06-10Publication date
2021-06-20Copyright date
2021ISSN
0888-3270Publisher version
Language
- en