| About this Abstract |
| Meeting |
2024 ASC Technical Conference, US-Japan Joint Symposium, D30 Meeting
|
| Symposium
|
2024 ASC Technical Conference, US-Japan Joint Symposium, D30 Meeting
|
| Presentation Title |
Deep Learning Models for Automatic Image Segmentation of Low Velocity Impact Damage in CFRP Composites |
| Author(s) |
Adrian Brelay, Olesya Zhupanska |
| On-Site Speaker (Planned) |
Adrian Brelay |
| Abstract Scope |
Low velocity impact damage in composites is one of the critical considerations in the design of damage tolerant composite structures. Particularly precarious situations arise when no apparent damage or barely visible damage at the impacted surface of the composite structure is accompanied by a significant reduction in structure’s strength, stiffness, and durability. Micro Computed Tomography (Micro-CT) emerged as one of the most advanced nondestructive evaluation methods that enables one to obtain high resolution 3D image data.
In this work, Machine Learning (ML) is used for the intent of Micro-CT image segmentation of low velocity impact damage in carbon fiber reinforced polymers (CFRP). Deep Learning models based on the U-Net Deep Learning architecture are trained and refined to provide better context on the accuracy of supervised ML algorithms as compared to unsupervised ML methods. The unsupervised ML algorithms relied on the statistical distances in conjunction with greyscale threshold intensity segmentation to isolate damage present in high resolution image data. Given the absence of standardization in image analysis of micro-CT data of composite materials, comparisons between supervised ML algorithms and unsupervised ML methods allow for investigation in the ability of Deep Learning models to accurately interpret various low velocity damage features. |
| Proceedings Inclusion? |
Definite: Post-meeting proceedings |