Segmentation of openings in non-crimp fabric using deep learning
Non-Crimp Fabrics (NCF) are widely used in industries like aerospace due to their lightweight and high-strength properties. Accurate characterization of openings in NCF is essential for maintaining material quality and performance. Traditional machine vision methods, such as histogram-based and edge...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Aeronautical and Astronautical Society of the Republic of China
2024
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/117868/ http://psasir.upm.edu.my/id/eprint/117868/1/117868.pdf |
| _version_ | 1848867365029150720 |
|---|---|
| author | Md Ali, Syaril Azrad Norhellme, Nornajme Ahmad Rashidi, Syazwan Mahmud Zuhudi, Nurul Zuhairah Abd Aziz, Noor Zuhaira |
| author_facet | Md Ali, Syaril Azrad Norhellme, Nornajme Ahmad Rashidi, Syazwan Mahmud Zuhudi, Nurul Zuhairah Abd Aziz, Noor Zuhaira |
| author_sort | Md Ali, Syaril Azrad |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | Non-Crimp Fabrics (NCF) are widely used in industries like aerospace due to their lightweight and high-strength properties. Accurate characterization of openings in NCF is essential for maintaining material quality and performance. Traditional machine vision methods, such as histogram-based and edge detection approaches, often struggle with complex NCF patterns and rely heavily on manual feature engineering. This study applies deep learning models U-Net, LinkNet, and DeepLabv3+ for the binary segmentation of openings in carbon fiber NCF. These models use a ResNet18 encoder, pre-trained on ImageNet, to automatically extract robust features and improve generalization. To address data limitations, we augmented the collected NCF dataset to increase variability. The results show that U-Net, LinkNet, and DeepLabv3+ achieved Intersection over Union (IoU) scores of 0.7419, 0.7228, and 0.7172, respectively, demonstrating their effectiveness in segmenting openings in NCF. This study presents a method that enhances segmentation accuracy and improves ability to generalize across varying data conditions, providing a more adaptable approach to NCF characterization with potential applications for other advanced materials. |
| first_indexed | 2025-11-15T14:35:20Z |
| format | Article |
| id | upm-117868 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T14:35:20Z |
| publishDate | 2024 |
| publisher | Aeronautical and Astronautical Society of the Republic of China |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1178682025-07-08T02:24:35Z http://psasir.upm.edu.my/id/eprint/117868/ Segmentation of openings in non-crimp fabric using deep learning Md Ali, Syaril Azrad Norhellme, Nornajme Ahmad Rashidi, Syazwan Mahmud Zuhudi, Nurul Zuhairah Abd Aziz, Noor Zuhaira Non-Crimp Fabrics (NCF) are widely used in industries like aerospace due to their lightweight and high-strength properties. Accurate characterization of openings in NCF is essential for maintaining material quality and performance. Traditional machine vision methods, such as histogram-based and edge detection approaches, often struggle with complex NCF patterns and rely heavily on manual feature engineering. This study applies deep learning models U-Net, LinkNet, and DeepLabv3+ for the binary segmentation of openings in carbon fiber NCF. These models use a ResNet18 encoder, pre-trained on ImageNet, to automatically extract robust features and improve generalization. To address data limitations, we augmented the collected NCF dataset to increase variability. The results show that U-Net, LinkNet, and DeepLabv3+ achieved Intersection over Union (IoU) scores of 0.7419, 0.7228, and 0.7172, respectively, demonstrating their effectiveness in segmenting openings in NCF. This study presents a method that enhances segmentation accuracy and improves ability to generalize across varying data conditions, providing a more adaptable approach to NCF characterization with potential applications for other advanced materials. Aeronautical and Astronautical Society of the Republic of China 2024 Article PeerReviewed text en http://psasir.upm.edu.my/id/eprint/117868/1/117868.pdf Md Ali, Syaril Azrad and Norhellme, Nornajme and Ahmad Rashidi, Syazwan and Mahmud Zuhudi, Nurul Zuhairah and Abd Aziz, Noor Zuhaira (2024) Segmentation of openings in non-crimp fabric using deep learning. Journal of Aeronautics, Astronautics and Aviation, 57 (3S). pp. 1-10. ISSN 1990-7710 https://www.airitilibrary.com/Article/Detail/P20140627004-N202504100011-00043 |
| spellingShingle | Md Ali, Syaril Azrad Norhellme, Nornajme Ahmad Rashidi, Syazwan Mahmud Zuhudi, Nurul Zuhairah Abd Aziz, Noor Zuhaira Segmentation of openings in non-crimp fabric using deep learning |
| title | Segmentation of openings in non-crimp fabric using deep learning |
| title_full | Segmentation of openings in non-crimp fabric using deep learning |
| title_fullStr | Segmentation of openings in non-crimp fabric using deep learning |
| title_full_unstemmed | Segmentation of openings in non-crimp fabric using deep learning |
| title_short | Segmentation of openings in non-crimp fabric using deep learning |
| title_sort | segmentation of openings in non-crimp fabric using deep learning |
| url | http://psasir.upm.edu.my/id/eprint/117868/ http://psasir.upm.edu.my/id/eprint/117868/ http://psasir.upm.edu.my/id/eprint/117868/1/117868.pdf |