Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network
Due to the increasing demand on road maintenance around the whole world, advanced techniques have been developed to automatically detect and segment pavement cracks. However, most of methods suffer from background noise or fail in fine crack segmentation. This paper proposes a generative adversarial...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-09-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123023003948 |
| _version_ | 1797682463116361728 |
|---|---|
| author | Zhihao Pan Stephen L.H. Lau Xu Yang Ningqun Guo Xin Wang |
| author_facet | Zhihao Pan Stephen L.H. Lau Xu Yang Ningqun Guo Xin Wang |
| author_sort | Zhihao Pan |
| collection | DOAJ |
| description | Due to the increasing demand on road maintenance around the whole world, advanced techniques have been developed to automatically detect and segment pavement cracks. However, most of methods suffer from background noise or fail in fine crack segmentation. This paper proposes a generative adversarial network (GAN)-based neural network named CrackSegAN to segment pavement cracks automatically. The generator of CrackSegAN generates segmentation results, while the discriminator trains the generator adversarially. A joint loss function is proposed to optimize the generator with sufficient gradients and mitigate the high class imbalance in pavement crack images. Elastic deformation data augmentation method is applied to force CrackSegAN to learn the transformation invariance. The proposed CrackSegAN reaches an average F1 score of 0.9780 on CrackForest dataset and 0.8412 on Crack500 dataset. Ablation study shows that the most prominent difference is made by the proposed joint loss function which increases the average F1 score by 8.98% on CrackForest dataset. Besides, the comparison between using different data augmentation strategies validates the effectiveness of elastic deformation. Overall, the proposed CrackSegAN increases the F1 score by 1.91% on CrackForest dataset and 1.01% on Crack500 compared with state-of-the-art methods. Qualitatively, CrackSegAN is more robust to background noises and segments cracks with more details. Moreover, the test on field data proves a better generalizability of CrackSegAN on unseen background noises. |
| first_indexed | 2024-03-12T00:00:05Z |
| format | Article |
| id | doaj.art-61b68e7c302344ab99d44ccdcdac22ce |
| institution | Directory Open Access Journal |
| issn | 2590-1230 |
| language | English |
| last_indexed | 2024-03-12T00:00:05Z |
| publishDate | 2023-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Engineering |
| spelling | doaj.art-61b68e7c302344ab99d44ccdcdac22ce2023-09-18T04:30:35ZengElsevierResults in Engineering2590-12302023-09-0119101267Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural networkZhihao Pan0Stephen L.H. Lau1Xu Yang2Ningqun Guo3Xin Wang4School of Engineering, Monash University Malaysia, Subang Jaya, 47500, MalaysiaSchool of Engineering, Monash University Malaysia, Subang Jaya, 47500, MalaysiaCollege of Future Transportation, Chang'an University, Xi'an 710064, China; Department of Civil Engineering, Monash University, Clayton, VIC, 3800, AustraliaSchool of Engineering, Monash University Malaysia, Subang Jaya, 47500, MalaysiaSchool of Engineering, Monash University Malaysia, Subang Jaya, 47500, Malaysia; Corresponding author.Due to the increasing demand on road maintenance around the whole world, advanced techniques have been developed to automatically detect and segment pavement cracks. However, most of methods suffer from background noise or fail in fine crack segmentation. This paper proposes a generative adversarial network (GAN)-based neural network named CrackSegAN to segment pavement cracks automatically. The generator of CrackSegAN generates segmentation results, while the discriminator trains the generator adversarially. A joint loss function is proposed to optimize the generator with sufficient gradients and mitigate the high class imbalance in pavement crack images. Elastic deformation data augmentation method is applied to force CrackSegAN to learn the transformation invariance. The proposed CrackSegAN reaches an average F1 score of 0.9780 on CrackForest dataset and 0.8412 on Crack500 dataset. Ablation study shows that the most prominent difference is made by the proposed joint loss function which increases the average F1 score by 8.98% on CrackForest dataset. Besides, the comparison between using different data augmentation strategies validates the effectiveness of elastic deformation. Overall, the proposed CrackSegAN increases the F1 score by 1.91% on CrackForest dataset and 1.01% on Crack500 compared with state-of-the-art methods. Qualitatively, CrackSegAN is more robust to background noises and segments cracks with more details. Moreover, the test on field data proves a better generalizability of CrackSegAN on unseen background noises.http://www.sciencedirect.com/science/article/pii/S2590123023003948Transportation safetyPavement crack segmentationDeep learningGenerative adversarial network (GAN)Fully convolutional network |
| spellingShingle | Zhihao Pan Stephen L.H. Lau Xu Yang Ningqun Guo Xin Wang Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network Results in Engineering Transportation safety Pavement crack segmentation Deep learning Generative adversarial network (GAN) Fully convolutional network |
| title | Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network |
| title_full | Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network |
| title_fullStr | Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network |
| title_full_unstemmed | Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network |
| title_short | Automatic pavement crack segmentation using a generative adversarial network (GAN)-based convolutional neural network |
| title_sort | automatic pavement crack segmentation using a generative adversarial network gan based convolutional neural network |
| topic | Transportation safety Pavement crack segmentation Deep learning Generative adversarial network (GAN) Fully convolutional network |
| url | http://www.sciencedirect.com/science/article/pii/S2590123023003948 |
| work_keys_str_mv | AT zhihaopan automaticpavementcracksegmentationusingagenerativeadversarialnetworkganbasedconvolutionalneuralnetwork AT stephenlhlau automaticpavementcracksegmentationusingagenerativeadversarialnetworkganbasedconvolutionalneuralnetwork AT xuyang automaticpavementcracksegmentationusingagenerativeadversarialnetworkganbasedconvolutionalneuralnetwork AT ningqunguo automaticpavementcracksegmentationusingagenerativeadversarialnetworkganbasedconvolutionalneuralnetwork AT xinwang automaticpavementcracksegmentationusingagenerativeadversarialnetworkganbasedconvolutionalneuralnetwork |