Journal ArticleDOI
Anomaly Detection and Classification in a Laser Powder Bed Additive Manufacturing Process using a Trained Computer Vision Algorithm
Luke Scime,Jack Beuth +1 more
Reads0
Chats0
TLDR
A computer vision algorithm is used to automatically detect and classify anomalies that occur during the powder spreading stage of the process, which has the potential to become a component of a real-time control system in an LPBF machine.Abstract:
Despite the rapid adoption of laser powder bed fusion (LPBF) Additive Manufacturing by industry, current processes remain largely open-loop, with limited real-time monitoring capabilities. While some machines offer powder bed visualization during builds, they lack automated analysis capability. This work presents an approach for in-situ monitoring and analysis of powder bed images with the potential to become a component of a real-time control system in an LPBF machine. Specifically, a computer vision algorithm is used to automatically detect and classify anomalies that occur during the powder spreading stage of the process. Anomaly detection and classification are implemented using an unsupervised machine learning algorithm, operating on a moderately-sized training database of image patches. The performance of the final algorithm is evaluated, and its usefulness as a standalone software package is demonstrated with several case studies.read more
Citations
More filters
Journal ArticleDOI
Powder Bed Defects Classification: An Industry Perspective
Francois Du Rand,Malan van Tonder,A. F. Van der Merwe,Olaf Diegel,Devon Hagedorn-Hansen,Ian Campbell,Ian Gibson +6 more
TL;DR: In this article , a defect severity classification matrix based on industry partner experience as well as published literature is used to autonomously classify defects, which can be used to make autonomous decisions regarding defects.
OtherDOI
Application of Machine Learning Algorithms and Models in 3D Printing
TL;DR: In this article , a neural network algorithm is attached to various parameters of the additive manufacturing chain, such as quality evaluation, in situ monitoring, and model design and the progress of these mechanisms is evaluated in this project.
Journal ArticleDOI
Heterogeneous sensor data fusion for multiscale, shape agnostic flaw detection in laser powder bed fusion additive manufacturing
Ben Bevans,Thomas G. Spears,Aniruddha Gaikwad,Alex Riensche,Ziyad M. Smoqi,Harold (Scott) Halliday,Prahalad K. Rao +6 more
TL;DR: In this article , a shape agnostic detection of multiscale flaws in laser powder bed fusion (LPBF) additive manufacturing using heterogenous in-situ sensor data was developed and applied.
Journal ArticleDOI
Powder Bed Monitoring Using Semantic Image Segmentation to Detect Failures during 3D Metal Printing
TL;DR: In this article , an Xception-style neural network was used to predict the powder and part areas in the metal additive manufacturing (AM) process and the segmentation result of every layer was compared to the reference layer regarding the area, centroids, and normalized area difference of each part.
References
More filters
Journal ArticleDOI
Deep learning
TL;DR: Deep learning is making major advances in solving problems that have resisted the best attempts of the artificial intelligence community for many years, and will have many more successes in the near future because it requires very little engineering by hand and can easily take advantage of increases in the amount of available computation and data.
Book
Deep Learning
TL;DR: Deep learning as mentioned in this paper is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts, and it is used in many applications such as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics, and videogames.
Journal ArticleDOI
An introduction to ROC analysis
TL;DR: The purpose of this article is to serve as an introduction to ROC graphs and as a guide for using them in research.
Proceedings ArticleDOI
Object recognition from local scale-invariant features
TL;DR: Experimental results show that robust object recognition can be achieved in cluttered partially occluded images with a computation time of under 2 seconds.
Proceedings Article
Visual categorization with bags of keypoints
TL;DR: This bag of keypoints method is based on vector quantization of affine invariant descriptors of image patches and shows that it is simple, computationally efficient and intrinsically invariant.