scispace - formally typeset
Proceedings ArticleDOI

Ultrasonic imaging of welds using boundary reflections

L. Hörchens, +2 more
- Vol. 1511, Iss: 1, pp 1051-1058
TLDR
In this paper, boundary reflections in the path from transmitter to receiver are used for the inspection of welds to determine the type, position, and size of defects, and these images can be combined into a single figure providing a section view of the weld.
Abstract
Ultrasonic imaging can be applied for the inspection of welds to determine the type, position, and size of defects. By including boundary reflections in the path from transmitter to receiver, defects in the weld can be imaged from different directions using both reflection and diffraction signals. These images can be combined into a single figure providing a section view of the weld. The geometrical configuration must be taken into account for proper alignment of the sub-images. These sub-images can be linked to conventional ultrasonic inspection techniques, such as the phased array tandem technique and time of flight diffraction.

read more

Citations
More filters
Journal ArticleDOI

Fusion of multi-view ultrasonic data for increased detection performance in non-destructive evaluation.

TL;DR: Fusion allows NDE capability to be extended with potential implications for the design and operation of engineering assets and fusion provides a means to substantially reduce operator burden.
Proceedings ArticleDOI

Deep Learning for Multi-View Ultrasonic Image Fusion

TL;DR: In this paper, a deep neural network (DNN) architecture is proposed to directly map all available data to a segmentation map while explicitly incorporating the DAS image formation for the different insonification paths as network layers.
Posted Content

Deep Learning for Multi-View Ultrasonic Image Fusion

TL;DR: In this paper, a deep neural network (DNN) architecture is proposed to directly map all available data to a segmentation map while explicitly incorporating the DAS image formation for the different insonification paths as network layers.
Related Papers (5)