UCAN: U-shaped context aggregation network for thin crack segmentation under topological constraints
Robotic Intelligence and Automation
ISSN: 2754-6969
Article publication date: 16 August 2024
Issue publication date: 29 August 2024
Abstract
Purpose
Thin cracks on the surface, such as those found in nuclear power plant concrete structures, are difficult to identify because they tend to be thin. This paper aims to design a novel segmentation network, called U-shaped contextual aggregation network (UCAN), for better recognition of weak cracks.
Design/methodology/approach
UCAN uses dilated convolutional layers with exponentially changing dilation rates to extract additional contextual features of thin cracks while preserving resolution. Furthermore, this paper has developed a topology-based loss function, called ℓcl Dice, which enhances the crack segmentation’s connectivity.
Findings
This paper generated five data sets with varying crack widths to evaluate the performance of multiple algorithms. The results show that the UCAN network proposed in this study achieves the highest F1-Score on thinner cracks. Additionally, training the UCAN network with the ℓcl Dice improves the F1-Scores compared to using the cross-entropy function alone. These findings demonstrate the effectiveness of the UCAN network and the value of incorporating the ℓcl Dice in crack segmentation tasks.
Originality/value
In this paper, an exponentially dilated convolutional layer is constructed to replace the commonly used pooling layer to improve the model receptive field. To address the challenge of preserving fracture connectivity segmentation, this paper introduces ℓcl Dice. This design enables UCAN to extract more contextual features while maintaining resolution, thus improving the crack segmentation performance. The proposed method is evaluated using extensive experiments where the results demonstrate the effectiveness of the algorithm.
Keywords
Acknowledgements
This work is supported in part by the National Natural Science Funds for Distinguished Young Scholar under Grant 62325307, in part by the National Natural Science Foundation of China under Grants 62072315, 62073225, 62176164, 62203134 and 62002239, in part by the Shenzhen Science and Technology Innovation Commission under Grants 20220809141216003, 20220809154139001, JCYJ20210324093808021 and JCYJ20220531102817040 and in part by the Scientific Instrument Developing Project of Shenzhen University under Grant 2023YQ019.
Citation
Chen, J., Zhu, G., Zhang, Y., Chen, Z., Huang, Q. and Li, J. (2024), "UCAN: U-shaped context aggregation network for thin crack segmentation under topological constraints", Robotic Intelligence and Automation, Vol. 44 No. 5, pp. 637-647. https://doi.org/10.1108/RIA-08-2023-0097
Publisher
:Emerald Publishing Limited
Copyright © 2024, Emerald Publishing Limited