Journal
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
Volume 23, Issue 8, Pages 10636-10645Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2021.3095167
Keywords
Fasteners; Inspection; Generative adversarial networks; Image synthesis; Feature extraction; Solid modeling; Rail transportation; Fastener inspection; image generation; GAN; classification model; deep learning
Categories
Funding
- National Natural Science Foundation of China [61771191, 61971182]
- Hunan Provincial Natural Science Foundation of China [2020JJ4213]
- Changsha City Science and Technology Department Funds [KQ2004007]
Ask authors/readers for more resources
This article proposes a method to improve the performance of deep learning-based defective fastener inspection by using a novel image generation technique. Experiments have shown that the generated defective fastener images have better quality and diversity compared to other methods, leading to significant improvement in the inspection model's performance.
This article aims to improve the performance of deep learning-based defective fastener inspection method. Due to the defective fasteners are insufficient and far less than the defect-free ones in real railway, it is difficult to train a robust fastener inspection model on such imbalanced dataset. In view of this problem, a novel image generation method called four-discriminator cycle-consistent adversarial network (FD-Cycle-GAN) is proposed to generate the defect fastener images using a large number of defect-free ones. Extensive experiments are conducted on the real fastener images and generated images. Experimental results demonstrate that the defect fastener images generated by our proposed method have better quality and richer diversity than those generated by other state-of-the-art methods. In addition, compared with the CNN-only baseline, the performance of the fastener inspection model trained on the expanded dataset containing the defect fastener images generated by FD-Cycle-GAN is improved significantly. The detection accuracy and relative IMP reach 93.25% and 21.59% respectively.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available