2026-10-12T12:32:59Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33988191 2026-09-24T17:34:01Z category_4 category_146 category_734 category_931 category_811 portal_5 item_type_3 month_year_09_2026

<p>Comparison of crack width calculation errors.</p> Xuwei Dong (25105161) Zhibing Yu (25105164) Jiashuo Yuan (25105167) Jinpeng Dai (25105170) Biochemistry Space Science Biological Sciences not elsewhere classified Information Systems not elsewhere classified Mathematical Sciences not elsewhere classified scale fusion attention scalable technical approach public dataset sdnet2018 proposed model achieves orthogonal skeleton method improving bridge operation effectively improves robustness bd detection algorithm local detailed features extremely low overhead experimental results show automatic crack detection achieve crack detection lightweight segmentation model feather sfaswinunet proposed feather sfaswinunet precise detection crack width segmentation results detailed modelling computational overhead feather convblock uav inspection study provides study proposes skip connections safety assessment recovery capability quantitatively analyse quantitative analysis miou value measurement errors mdice value maintenance efficiency maintenance decisions highly subjective feature distortion fcb compensates fcb ). complex backgrounds branch structure 57 %. <div><p>Bridge cracks are among the primary factors affecting structural safety and durability. Intelligent and precise detection and quantification of cracks are important in improving bridge operation and maintenance efficiency and safety assessment. Traditional manual detection methods are inefficient and highly subjective, while existing deep learning methods still have shortcomings in terms of computational overhead and detailed modelling. Therefore, this study proposes a high-precision and lightweight crack semantic segmentation network—Feather SFASwinUnet. The model introduces two new modules: Scale Fusion Attention (SFA) and Feather ConvBlock (FCB). SFA effectively fuses global semantic information and local detailed features through a dual-branch structure, which improves spatial structure modelling capabilities. Meanwhile, FCB compensates for feature distortion with extremely low overhead in skip connections, which enhances the recovery capability of fine-grained cracks. Experimental results show that, on the public dataset SDNET2018, the proposed model achieves an mIoU value of 81.25% and an mDice value of 88.57%. It significantly outperforms typical methods such as Unet, DeepLabV3+ and SwinUnet while maintaining low computational load. Furthermore, the YOLO11-BD detection algorithm is combined to achieve crack detection and segmentation under complex backgrounds, which effectively improves robustness. Lastly, the orthogonal skeleton method is used to quantitatively analyse the segmentation results. As a result, the measurement errors of crack width and length are controlled within 5%, which provides reliable technical support for bridge structural health assessment and maintenance decisions. The Feather SFASwinUnet proposed in this study provides an efficient and scalable technical approach for automatic crack detection and quantitative analysis of bridges under UAV inspection. In addition, it has high engineering application potential.</p></div> 2026-09-24T17:33:44Z Dataset Dataset 10.1371/journal.pone.0358865.t008 https://figshare.com/articles/dataset/_p_Comparison_of_crack_width_calculation_errors_p_/33988191 CC BY 4.0