| About this Abstract |
| Meeting |
NUMISHEET 2022: The 12th International Conference on Numerical Simulation of 3D Sheet Metal Forming Processes
|
| Symposium
|
NUMISHEET 2022
|
| Presentation Title |
Deep Learning Based Defect Inspection in Sheet Metal Stamping Parts |
| Author(s) |
Aru Ranjan Singh, Thomas Bashford-Rogers, Sumit Hazra, Kurt Debattista |
| On-Site Speaker (Planned) |
Aru Ranjan Singh |
| Abstract Scope |
Defects inspection is a crucial step in sheet metal stamping manufacturing. However, current inspection methods largely consist of visual inspection by trained operatives are unreliable and prone to error. Computer vision techniques have the potential advantages of utilising low-cost hardware to enable accurate classification, particularly through new techniques such as deep learning.
At present vision-based deep learning models for sheet material are limited to detecting defects in flat surfaces. This research proposes a practical deep learning approach for the classification of realistically formed sheet metal stamping components and suggests a route towards reliable and automated inspections in sheet metal stamping.
This study used Resnet18 a state-of-the-art deep learning model to classify split defects in "Nakajima" stamped components. The model was able to achieve a 99.84% accuracy on the validation set which implies that this technique could be suitable for automated defect detection on stamped metal parts. |
| Proceedings Inclusion? |
Definite: At-meeting proceedings |