About this Abstract |
Meeting |
2022 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2022)
|
Symposium
|
2022 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2022)
|
Presentation Title |
Convolutional Autoencoder for Image Denoising AM Process Monitoring |
Author(s) |
Jaison Mneyergi, Richard Anarfi, Benjamin Kwapong, Kenneth Fletcher, Todd Sparks, Aaron Flood |
On-Site Speaker (Planned) |
Jaison Mneyergi |
Abstract Scope |
To properly monitor the directed energy deposition (DED) additive manufacturing process, real time pictures/videos are sometimes transmitted to the human experts or engineered algorithms to determine whether the process is achieving its desired goal as at that point. Image noise from in-camera and in-process sources can obscure some features and make it difficult to ascertain whether the manufacturing process is proceeding as expected at that time. In this paper, we propose to use a computer vision approach to train a denoising convolutional autoencoder that learns to recreate clean images from noisy ones. It does this by learning the important features from the images and using those features to reconstruct the images. We conduct experiments with simulated images from a powder-based DED process. The results from our experiments show that our model successfully removes varying forms of noise from the images. |
Proceedings Inclusion? |
Definite: Post-meeting proceedings |