GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation
Pavement cracks are the primary type of distress that cause road damage, and deep-learning-based pavement crack segmentation is a critical technology for current pavement maintenance and management. To address the issues of segmentation discontinuity and poor performance in the segmentation of irreg...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/15/3348 |
| _version_ | 1797586880697466880 |
|---|---|
| author | Yawei Qi Fang Wan Guangbo Lei Wei Liu Li Xu Zhiwei Ye Wen Zhou |
| author_facet | Yawei Qi Fang Wan Guangbo Lei Wei Liu Li Xu Zhiwei Ye Wen Zhou |
| author_sort | Yawei Qi |
| collection | DOAJ |
| description | Pavement cracks are the primary type of distress that cause road damage, and deep-learning-based pavement crack segmentation is a critical technology for current pavement maintenance and management. To address the issues of segmentation discontinuity and poor performance in the segmentation of irregular cracks faced by current semantic segmentation models, this paper proposes an irregular pavement crack segmentation method based on multi-scale convolutional attention aggregation. In this approach, GhostNet is first introduced as the model backbone network for reducing parameter count, with dynamic convolution enhancing GhostNet’s feature extraction capability. Next, a multi-scale convolutional attention aggregation module is proposed to cause the model to focus more on crack features and thus improve the segmentation effect on irregular cracks. Finally, a progressive up-sampling structure is used to enrich the feature information by gradually fusing feature maps of different depths to enhance the continuity of segmentation results. The experimental results on the HGCrack dataset show that GMDNet has a lighter model structure and higher segmentation accuracy than the mainstream semantic segmentation algorithms, achieving 75.16% of MIoU and 84.43% of F1 score, with only 7.67 M parameters. Therefore, the GMDNet proposed in this paper can accurately and efficiently segment irregular cracks on pavements that are more suitable for pavement crack segmentation scenarios in practical applications. |
| first_indexed | 2024-03-11T00:28:26Z |
| format | Article |
| id | doaj.art-710376bc80c6402cbe2a6ab472b90f18 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T00:28:26Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-710376bc80c6402cbe2a6ab472b90f182023-11-18T22:49:47ZengMDPI AGElectronics2079-92922023-08-011215334810.3390/electronics12153348GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention AggregationYawei Qi0Fang Wan1Guangbo Lei2Wei Liu3Li Xu4Zhiwei Ye5Wen Zhou6School of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaPavement cracks are the primary type of distress that cause road damage, and deep-learning-based pavement crack segmentation is a critical technology for current pavement maintenance and management. To address the issues of segmentation discontinuity and poor performance in the segmentation of irregular cracks faced by current semantic segmentation models, this paper proposes an irregular pavement crack segmentation method based on multi-scale convolutional attention aggregation. In this approach, GhostNet is first introduced as the model backbone network for reducing parameter count, with dynamic convolution enhancing GhostNet’s feature extraction capability. Next, a multi-scale convolutional attention aggregation module is proposed to cause the model to focus more on crack features and thus improve the segmentation effect on irregular cracks. Finally, a progressive up-sampling structure is used to enrich the feature information by gradually fusing feature maps of different depths to enhance the continuity of segmentation results. The experimental results on the HGCrack dataset show that GMDNet has a lighter model structure and higher segmentation accuracy than the mainstream semantic segmentation algorithms, achieving 75.16% of MIoU and 84.43% of F1 score, with only 7.67 M parameters. Therefore, the GMDNet proposed in this paper can accurately and efficiently segment irregular cracks on pavements that are more suitable for pavement crack segmentation scenarios in practical applications.https://www.mdpi.com/2079-9292/12/15/3348pavement crackingsemantic segmentationlightweight modeldynamic convolutionmulti-scale convolutional attention |
| spellingShingle | Yawei Qi Fang Wan Guangbo Lei Wei Liu Li Xu Zhiwei Ye Wen Zhou GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation Electronics pavement cracking semantic segmentation lightweight model dynamic convolution multi-scale convolutional attention |
| title | GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation |
| title_full | GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation |
| title_fullStr | GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation |
| title_full_unstemmed | GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation |
| title_short | GMDNet: An Irregular Pavement Crack Segmentation Method Based on Multi-Scale Convolutional Attention Aggregation |
| title_sort | gmdnet an irregular pavement crack segmentation method based on multi scale convolutional attention aggregation |
| topic | pavement cracking semantic segmentation lightweight model dynamic convolution multi-scale convolutional attention |
| url | https://www.mdpi.com/2079-9292/12/15/3348 |
| work_keys_str_mv | AT yaweiqi gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation AT fangwan gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation AT guangbolei gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation AT weiliu gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation AT lixu gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation AT zhiweiye gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation AT wenzhou gmdnetanirregularpavementcracksegmentationmethodbasedonmultiscaleconvolutionalattentionaggregation |