A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Asymmetric dual-decoder-U-Net for pavement crack semantic segmentation
https://orcid.org/0000-0002-8920-7635
https://orcid.org/0000-0002-4457-4407
https://orcid.org/0000-0002-3200-4781
https://orcid.org/0000-0001-8396-5710
Abstract Accurate pavement crack segmentation is crucial for civil engineering and infrastructure maintenance. To address the challenge of imbalanced data resulting from the prevalence of non-crack pixels, this research seeks to improve the quality of pavement crack segmentation, particularly for thick and tiny cracks. This paper presents an Asymmetric Dual-Decoder-U-Net (ADDU-Net) model, which involves constructing an asymmetric dual decoder with a dual attention module to better capture the features of both thick and tiny cracks under diverse environmental conditions. Through evaluation with images from four benchmark datasets, the ADDU-Net model demonstrates its effectiveness and robustness in accurately segmenting various types of cracks. This segmentation model shows significant potential for improving crack segmentation in industrial applications.
Graphical abstract Display Omitted
Highlights A pavement crack semantic segmentation model based on asymmetric dual decoders. A dual attention module to boost feature map representations from input images. A hybrid loss function is developed to tackle the problem of class imbalance. A residual refinement module is proposed to optimize crack boundaries.
Asymmetric dual-decoder-U-Net for pavement crack semantic segmentation
https://orcid.org/0000-0002-8920-7635
https://orcid.org/0000-0002-4457-4407
https://orcid.org/0000-0002-3200-4781
https://orcid.org/0000-0001-8396-5710
Abstract Accurate pavement crack segmentation is crucial for civil engineering and infrastructure maintenance. To address the challenge of imbalanced data resulting from the prevalence of non-crack pixels, this research seeks to improve the quality of pavement crack segmentation, particularly for thick and tiny cracks. This paper presents an Asymmetric Dual-Decoder-U-Net (ADDU-Net) model, which involves constructing an asymmetric dual decoder with a dual attention module to better capture the features of both thick and tiny cracks under diverse environmental conditions. Through evaluation with images from four benchmark datasets, the ADDU-Net model demonstrates its effectiveness and robustness in accurately segmenting various types of cracks. This segmentation model shows significant potential for improving crack segmentation in industrial applications.
Graphical abstract Display Omitted
Highlights A pavement crack semantic segmentation model based on asymmetric dual decoders. A dual attention module to boost feature map representations from input images. A hybrid loss function is developed to tackle the problem of class imbalance. A residual refinement module is proposed to optimize crack boundaries.
Asymmetric dual-decoder-U-Net for pavement crack semantic segmentation
https://orcid.org/0000-0002-8920-7635
https://orcid.org/0000-0002-4457-4407
https://orcid.org/0000-0002-3200-4781
https://orcid.org/0000-0001-8396-5710
Abstract Accurate pavement crack segmentation is crucial for civil engineering and infrastructure maintenance. To address the challenge of imbalanced data resulting from the prevalence of non-crack pixels, this research seeks to improve the quality of pavement crack segmentation, particularly for thick and tiny cracks. This paper presents an Asymmetric Dual-Decoder-U-Net (ADDU-Net) model, which involves constructing an asymmetric dual decoder with a dual attention module to better capture the features of both thick and tiny cracks under diverse environmental conditions. Through evaluation with images from four benchmark datasets, the ADDU-Net model demonstrates its effectiveness and robustness in accurately segmenting various types of cracks. This segmentation model shows significant potential for improving crack segmentation in industrial applications.
Graphical abstract Display Omitted
Highlights A pavement crack semantic segmentation model based on asymmetric dual decoders. A dual attention module to boost feature map representations from input images. A hybrid loss function is developed to tackle the problem of class imbalance. A residual refinement module is proposed to optimize crack boundaries.
Asymmetric dual-decoder-U-Net for pavement crack semantic segmentation
Al-Huda, Zaid (author) / Peng, Bo (author) / Algburi, Riyadh Nazar Ali (author) / Al-antari, Mugahed A. (author) / AL-Jarazi, Rabea (author) / Al-maqtari, Omar (author) / Zhai, Donghai (author)
2023-10-16
Article (Journal)
Electronic Resource
English
A complex scene pavement crack semantic segmentation method based on dual-stream framework
| Taylor & Francis Verlag | 2023
| Taylor & Francis Verlag | 2023
Semi-supervised semantic segmentation using cross-consistency training for pavement crack detection
| Taylor & Francis Verlag | 2024