Please use this identifier to cite or link to this item:
Title: Anomaly detection in aerosol jet printing via computer vision & machine learning
Authors: Ong, Alvin Wei Siang
Keywords: Engineering::Manufacturing::Quality control
Issue Date: 2022
Publisher: Nanyang Technological University
Source: Ong, A. W. S. (2022). Anomaly detection in aerosol jet printing via computer vision & machine learning. Final Year Project (FYP), Nanyang Technological University, Singapore.
Project: C083
Abstract: A gamut of industries is adopting additive manufacturing processes and the prevalence of these technologies are growing sharply. There is ongoing research which mainly focuses on the improvement and advancement of AM processes. Regardless, several drawbacks with regards to the variability in printing and printing quality parts are some of the known issues in AM processes. The frequency of occurrence of defects could often lead to such detrimental occurrence. This paper aims to develop and implement an in-situ monitoring system on Aerosol Jet Printing (AJP) to detect in real-time with the use of an Object Detection Model and Computer Vision. Image data of six classes of defects were collected for model training. An object detection model was selected, trained, and evaluated against several metrics. The selected model gave a classification accuracy of 84.7% and an inference speed of 69.93 FPS. Although the results computed from the experiment were far from expectation, additional discussions will be supplemented to look into the improvements that can be made to allow for in-situ monitoring of AJP process.
Schools: School of Mechanical and Aerospace Engineering 
Research Centres: Singapore Centre for 3D Printing 
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:MAE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
C083 Anomaly Detection in Aerosol Jet Printing via Computer Vision & Machine Learning.pdf
  Restricted Access
2.71 MBAdobe PDFView/Open

Page view(s)

Updated on Dec 8, 2023


Updated on Dec 8, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.