3D U-Net-based segmentation of voids and delamination in composite materials using micro-CT data

Key Points

• Effective segmentation of voids and delamination is possible using 3D U-Net, enhancing material analysis.
• The model achieved an accuracy of over 90% in identifying voids in micro-CT data images.
• Using micro-CT data allows for detailed insights into the internal structures of composite materials.
• This approach highlights the potential for improved quality control in manufacturing processes.