Related Experiment Video
Updated: Jun 26, 2026
11:34
Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
Published on: May 15, 2017
11.1K
Deep-Learning-Based Segmentation of Keyhole in In-Situ X-ray Imaging of Laser Powder Bed Fusion.
William Dong1, Jason Lian2, Chengpo Yan2
1Department of Mechanical Engineering, University of Wisconsin-Madison, Madison, WI 53706, USA.
Materials (Basel, Switzerland)
|January 26, 2024
Summary
An automated tool uses deep learning to segment keyhole shapes from X-ray images in laser powder bed fusion. This computer vision approach accurately identifies keyhole regions, improving defect analysis in metal additive manufacturing.
Area of Science:
- Materials Science
- Additive Manufacturing
- Computer Vision
Background:
- Keyholes, gaseous cavities formed during laser powder bed fusion (LPBF), significantly influence defect formation and product quality.
- In-situ X-ray imaging captures keyhole dynamics, but manual labeling of keyhole shapes for machine learning analysis is inefficient and error-prone.
- Automating keyhole shape identification is crucial for scalable analysis and correlating morphology with defects.
Purpose of the Study:
- To develop and present a deep-learning-based computer vision tool for automatic segmentation of keyhole shapes from X-ray images.
- To enable efficient and scalable analysis of keyhole dynamics in LPBF processes.
- To facilitate the use of keyhole shapes as input for machine learning models to predict defects.
Main Methods:
- A computer vision pipeline integrating a filtering method and the BASNet deep learning model was implemented.
- The tool performs semantic segmentation to extract keyhole morphologies from X-ray image streams.
- The approach was trained and tested on X-ray images of Al6061 and AliSi10Mg alloys.
Main Results:
- The automated tool achieved high accuracy: 91.24% for keyhole area and 92.81% for boundary shape.
- Performance was validated across various test dataset conditions.
- The system utilized 300 training images/labels and 100 testing images per trial.
Conclusions:
- The presented deep learning tool effectively automates the segmentation of keyhole shapes from X-ray images in LPBF.
- This automated approach offers a scalable and accurate solution for analyzing keyhole dynamics and defect prediction.
- The tool can be directly applied or retrained for large-scale image dataset analysis in additive manufacturing research.
Related Concept Videos
01:16Confocal Fluorescence Microscopy
Confocal microscopy is an advanced microscopic technique. The prime advantage of the confocal microscope over other microscopy techniques is its ability to block the out-of-focus light from the illuminated samples using pinholes. It is widely used with fluorescence optics to obtain high-resolution, sharp contrast images. Unlike optical microscopes, confocal microscopes use a focused beam of light laser to scan the entire sample surface at different z-planes. These microscopes are, therefore,...
01:24X-ray Imaging
German physicist Wilhelm Röntgen (1845–1923) was experimenting with electrical current when he discovered that a mysterious and invisible "ray" would pass through his flesh but leave an outline of his bones on a screen coated with a metal compound. In 1895, Röntgen made the first durable record of the internal parts of a living human: an "X-ray" image (as it came to be called) of his wife’s hand. Scientists worldwide quickly began their own experiments with X-rays, and by 1900, X-ray was widely...