Online quality inspection using Bayesian classification in powder-bed additive manufacturing from high-resolution visual camera images

Masoumeh Aminzadeh () and Thomas R. Kurfess
Additional contact information
Masoumeh Aminzadeh: Georgia Institute of Technology
Thomas R. Kurfess: Georgia Institute of Technology

Journal of Intelligent Manufacturing, 2019, vol. 30, issue 6, No 12, 2505-2523

Abstract: Abstract Despite their advances and numerous benefits, metal powder-bed additive manufacturing (AM) processes still suffer from the high chances of defect formation and a need for improved quality. This work develops an online monitoring system for quality of fusion and defect formation in every layer of the laser powder-bed fusion process using computer vision and Bayesian inference. An imaging setup is developed that for the first time allows capturing in-situ (during the build) images from every layer that visualize detailed layer defects and porosity. A database of camera images from every layer of AM parts made with various part quality was created that is the first visual labeled dataset from in-situ visual images of the powder-bed AM (also visualizing detailed layer features). The dataset is used in training-based classification to detect layers or sub-regions of the layer with low quality of fusion or defects. Features are carefully selected based on physical intuition into the process and extracted from the images of the various types of builds. A Bayesian classifier is developed and trained to classify the quality of the build that signifies the defective and unacceptable build layers or regions. The results can be used for quasi-real-time (layer-wise) process control, further process decisions, or corrective actions.

Keywords: Additive manufacturing; Online quality inspection; In-situ defect detection; Bayesian inference; Supervised learning; Feature-based classification; Computer vision; Metal powder-bed additive manufacturing; Laser powder-bed fusion; 3D printing (search for similar items in EconPapers)
Date: 2019
References: View complete reference list from CitEc
Citations: View citations in EconPapers (10)

Downloads: (external link)
http://link.springer.com/10.1007/s10845-018-1412-0 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:30:y:2019:i:6:d:10.1007_s10845-018-1412-0

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-018-1412-0

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().