File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1002/stc.2850
- Scopus: eid_2-s2.0-85116439841
- WOS: WOS:000704295800001
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Deep convolutional neural networks for semantic segmentation of cracks
Title | Deep convolutional neural networks for semantic segmentation of cracks |
---|---|
Authors | |
Keywords | computer vision convolutional neural network crack deep learning encoder–decoder semantic segmentation |
Issue Date | 2022 |
Citation | Structural Control and Health Monitoring, 2022, v. 29, n. 1, article no. e2850 How to Cite? |
Abstract | A large crack detection dataset of 2446 manually labeled images is established to cover a wide range of noise and to evaluate the performance of end-to-end deep convolutional networks in detecting cracking. Five state-of-the-art end-to-end deep computer vision architectures for semantic segmentation are trained and evaluated, including Fully Convolutional Network (FCN), Global Convolutional Network (GCN), Pyramid Scene Parsing Network (PSPNet), UPerNet, and DeepLabv3+. For the backbones, the VGG, ResNet, and DenseNet are adopted. Based on the comparison of test set metrics, DeepLabv3+ with the ResNet101 backbone achieved the highest IoU of 0.6298, the highest recall of 0.6834, and the highest F1 score of 0.7732. The influence of database choice and image noise on crack detection performance is reported. Based on the comparison of predicted images, UperNet with ResNet101 backbone shows the highest performance for images with shadings, while DeepLabv3+ with ResNet101 backbone shows the best performance for images with blemishes. The research outcome can provide reference for the application of fast and accurate detection of cracks in civil engineering. |
Persistent Identifier | http://hdl.handle.net/10722/326300 |
ISSN | 2023 Impact Factor: 4.6 2023 SCImago Journal Rankings: 1.349 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Wang, Jia Ji | - |
dc.contributor.author | Liu, Yu Fei | - |
dc.contributor.author | Nie, Xin | - |
dc.contributor.author | Mo, Y. L. | - |
dc.date.accessioned | 2023-03-09T09:59:36Z | - |
dc.date.available | 2023-03-09T09:59:36Z | - |
dc.date.issued | 2022 | - |
dc.identifier.citation | Structural Control and Health Monitoring, 2022, v. 29, n. 1, article no. e2850 | - |
dc.identifier.issn | 1545-2255 | - |
dc.identifier.uri | http://hdl.handle.net/10722/326300 | - |
dc.description.abstract | A large crack detection dataset of 2446 manually labeled images is established to cover a wide range of noise and to evaluate the performance of end-to-end deep convolutional networks in detecting cracking. Five state-of-the-art end-to-end deep computer vision architectures for semantic segmentation are trained and evaluated, including Fully Convolutional Network (FCN), Global Convolutional Network (GCN), Pyramid Scene Parsing Network (PSPNet), UPerNet, and DeepLabv3+. For the backbones, the VGG, ResNet, and DenseNet are adopted. Based on the comparison of test set metrics, DeepLabv3+ with the ResNet101 backbone achieved the highest IoU of 0.6298, the highest recall of 0.6834, and the highest F1 score of 0.7732. The influence of database choice and image noise on crack detection performance is reported. Based on the comparison of predicted images, UperNet with ResNet101 backbone shows the highest performance for images with shadings, while DeepLabv3+ with ResNet101 backbone shows the best performance for images with blemishes. The research outcome can provide reference for the application of fast and accurate detection of cracks in civil engineering. | - |
dc.language | eng | - |
dc.relation.ispartof | Structural Control and Health Monitoring | - |
dc.subject | computer vision | - |
dc.subject | convolutional neural network | - |
dc.subject | crack | - |
dc.subject | deep learning | - |
dc.subject | encoder–decoder | - |
dc.subject | semantic segmentation | - |
dc.title | Deep convolutional neural networks for semantic segmentation of cracks | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1002/stc.2850 | - |
dc.identifier.scopus | eid_2-s2.0-85116439841 | - |
dc.identifier.volume | 29 | - |
dc.identifier.issue | 1 | - |
dc.identifier.spage | article no. e2850 | - |
dc.identifier.epage | article no. e2850 | - |
dc.identifier.eissn | 1545-2263 | - |
dc.identifier.isi | WOS:000704295800001 | - |