Edge-Enhanced CrackNet for Underwater Crack Detection in Concrete Dams
Source abstract
Abstract
Underwater crack detection in dam structures is of significant importance to ensure structural safety, assess operational conditions, and prevent potential disasters. Traditional crack detection methods face various limitations when applied to underwater environments, particularly in high dam underwater environments where image quality is influenced by factors such as water flow disturbances, light diffraction effects, and low contrast, making it difficult for conventional methods to accurately extract crack features. This study proposes a dual-stage underwater crack detection method based on Cycle-GAN and YOLOv11 called Edge-Enhanced Underwater CrackNet (E2UCN) to overcome the limitations of existing image enhancement methods in retaining crack details and improving detection accuracy. First, underwater concrete crack images were collected using an underwater remotely operated vehicle (ROV), and various complex underwater environments were simulated to construct a test dataset. Then, an improved Cycle-GAN image style transfer method was used to enhance the underwater images. Unlike conventional GAN-based underwater image enhancement methods that focus on global visual quality, our model specifically constrains edge preservation and high-frequency crack textures, providing a novel solution tailored for crack detection tasks. Subsequently, the YOLOv11 model was employed to perform object detection on the enhanced underwater crack images, effectively extracting crack features and achieving high-precision crack detection. The experimental results show that the proposed method significantly outperforms traditional methods in terms of crack detection accuracy, edge clarity, and adaptability to complex backgrounds, effectively improving underwater crack detection accuracy (precision = 0.995, F1 = 0.99762, mAP@0.5 = 0.995, and mAP@0.5:0.95 = 0.736) and providing a feasible technological solution for intelligent inspection of high dam underwater cracks.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Infrastructure Maintenance and Monitoring · Dam Engineering and Safety · Geotechnical Engineering and Underground Structures
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Nanjing Hydraulic Research Institute, Nanjing 210029, China
NHRI R&D Tech Group Co., Ltd., Nanjing 210029, China
Dam Safety Management Center of the Ministry of Water Resources, Nanjing 210029, China
Funding
National Key R&D Program of China · National Natural Science Foundation of China · National Natural Science Foundation of China—Joint Fund · Jiangsu Province Youth Science and Technology Talent Support Program · Scientific Research Fund of Nanjing Hydraulic Research Institute · Technology Talent and Platform Program of Yunnan Province · Science and Technology Projects of China Huaneng Group
References (43)
Show source-provided references
1. Gong, X.N., Zhang, C.S., and Jia, J.S. (2021). Dam Danger Assessment and Risk Reinforcement Technology, China Construction Industry Press. (In Chinese).
2. Underwater defect detection and diagnosis assessment for high dams: Current status and challenges
3. Underwater image enhancement: A comprehensive review, recent trends, challenges and applications10.1007/s10462-021-10025-z ↗
4. Akkaynak, D., and Treibitz, T. (2019, January 15–20). Sea-thru: A Method for Removing Water from Underwater Images. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00178 ↗
5. Adaptive histogram equalization and its variations10.1016/S0734-189X(87)80186-X ↗
6. Adaptive homomorphic filtering algorithm for low illumination image enhancement
7. Adaptively Dictionary Construction for Hyperspectral Target Detection
8. Dam early warning model based on structural anomaly identification and dynamic effect variables selection10.1016/j.istruc.2025.108507 ↗
9. Distribution of strong input energy in base-isolated structures with complex nonlinearity: A parametric assessment10.1108/MMMS-08-2022-0158 ↗
10. Developing machine learning models for identifying the failure potential of fire-exposed FRP-strengthened concrete beams10.1007/s13369-024-09497-2 ↗
11. WaterGAN: Unsupervised generative network to enable real-time color correction of monocular underwater images
12. Naik, A., Swarnakar, A., and Mittal, K. (2021, January 2–9). Shallow-uwnet: Compressed model for underwater image enhancement (student abstract). Proceedings of the AAAI Conference on Artificial Intelligence, Online.10.1609/aaai.v35i18.17923 ↗
13. Underwater color correction network with knowledge transfer10.1109/TMM.2024.3374598 ↗
14. Fu, Z., Wang, W., Huang, Y., Ding, X., and Ma, K. (2022). Uncertainty inspired underwater image enhancement. Computer Vision—ECCV 2022, 17th European Conference, Tel Aviv, Israel, 23–27 October 2022, Springer Nature.10.1007/978-3-031-19797-0_27 ↗
15. Underwater Image Enhancement Based on Multi-scale Attention and Contrastive Learning
16. Du, F., Wang, H., Yao, H., and Chen, X. (2024). Domain-Adaptive Underwater Image Enhancement Algorithm. Comput. Mod., 55–60. (In Chinese).
17. Li, J., Skinner, K., Eustice, R.M., and Johnson-Roberson, M. (2017). WaterGAN: Unsupervised Generative Network to Enable Real-time Color Correction of Monocular Underwater Images. arXiv.10.1109/LRA.2017.2730363 ↗
18. Fast underwater image enhancement for improved visual perception10.1109/LRA.2020.2974710 ↗
19. Electronical imaging of structural concrete: Investigation by Bayesian stationary wavelet field
20. Gao, Y., Wang, Y., and Zhou, D. (2010, January 7–9). The Image Recognition, Automatic Measurement and Seam Tracking Technology in Arc Welding Process. Proceedings of the 2010 8th World Congress on Intelligent Control and Automation, Jinan, China.
21. CrackYOLO: Towards efficient dam crack detection for underwater scenes10.1007/s10044-024-01310-y ↗
22. Mao, Y., Ping, P., Chen, J., and Chen, H. (2020, January 25–28). Crack Detection with Multi-task Enhanced Faster R-CNN Model. Proceedings of the IEEE International Conference on Image Processing, Online.10.1109/BigDataService49289.2020.00038 ↗
23. Real-Time Crack Detection Method for Concrete Dams Based on Object Detection
24. Zheng, Z., Wang, P., Liu, W., Li, J., Ye, R., and Ren, D. (2020, January 7–12). Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression. Proceedings of the the AAAI Conference on Artificial Intelligence, New York, NY, USA.10.1609/aaai.v34i07.6999 ↗
25. YOLOv8-UC: An Improved YOLOv8-Based Underwater Object Detection Algorithm10.1109/ACCESS.2024.3496925 ↗
26. Liu, C., Li, H., Wang, S., Zhu, M., Wang, D., Fan, X., and Wang, Z. (2021, January 5–9). A dataset and benchmark of underwater object detection for robot picking. Proceedings of the 2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), Shenzhen, China.10.1109/ICMEW53276.2021.9455997 ↗
27. Zhu, J.-Y., Park, T., Isola, P., and Efros, A.A. (2017, January 22–29). Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy.10.1109/ICCV.2017.244 ↗
28. Zhang, Y., Wu, K., and Fang, K. (2024). Comprehensive Performance Evaluation of YOLOv11: Advancements, Benchmarks, and Real-World Applications. arXiv.
29. AttentionGAN: Unpaired Image-to-Image Translation Using Attention-Guided Generative Adversarial Networks10.1109/TNNLS.2021.3105725 ↗
30. Semi—Supervised image super-resolution with attention CycleGAN10.1049/ipr2.12401 ↗
31. Wang, X., Yu, K., Wu, S., Gu, J., Liu, Y., Dong, C., Qiao, Y., and Change Loy, C. (2018, January 8–14). Esrgan: Enhanced super-resolution generative adversarial networks. Proceedings of the European Conference on Computer Vision (ECCV) Workshops, Munich, Germany.10.1007/978-3-030-11021-5_5 ↗
32. Wei, Z., Huang, Y., Chen, Y., Zheng, C., and Gao, J. (2023). A-ESRGAN: Training real-world blind super-resolution with attention U-Net Discriminators. PRICAI 2023: Trends in Artificial Intelligence, 20th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2023, Jakarta, Indonesia, 15–19 November 2023, Springer Nature.10.1007/978-981-99-7025-4_2 ↗
33. Design of an image edge detection filter using the Sobel operator10.1109/4.996 ↗
34. An algorithm for the machine calculation of complex Fourier series10.1090/S0025-5718-1965-0178586-1 ↗
35. Gonzalez, R.C., and Woods, R.E. (2008). Digital Image Processing, Pearson. [3rd ed.].
36. Jiang, L., Dai, B., Wu, W., and Loy, C.C. (2021, January 6–14). Focal Frequency Loss for Image Reconstruction and Synthesis. Proceedings of the 2021 Conference on Neural Information Processing Systems, Online.10.1109/ICCV48922.2021.01366 ↗
37. Kim, M.W., and Cho, N.I. (2023, January 2–7). Wavelet-Domain High-Frequency Loss for Perceptual Quality. Proceedings of the 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, USA.
38. Wu, J., Zhang, G., and Fan, Y. (2024). LM-CycleGAN: Improving Underwater Image Quality Through Learned Perceptual Image Patch Similarity and Multi-Scale Adaptive Fusion Attention. Sensors, 24.10.3390/s24237425 ↗
39. Enhancement of underwater dam crack images using a multi-feature CycleGAN10.1016/j.autcon.2024.105727 ↗
40. Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness and Correlation
41. The PASCAL Visual Object Classes (VOC) Challenge10.1007/s11263-009-0275-4 ↗
42. Yacouby, R., and Axman, D. (, January November). Probabilistic extension of precision, recall, and f1 score for more thorough evaluation of classification models. Proceedings of the First Workshop on Evaluation and Comparison of NLP Systems, Online. Available online: https://aclanthology.org/2020.eval4nlp-1.9/.10.18653/v1/2020.eval4nlp-1.9 ↗
43. Zhang, W., Wu, X., and Xiang, Y. (2025, January 16–23). Edge enhanced underwater CrackNet for dam underwater crack detection. Proceedings of the International Symposium “Common Challenges, Shared Future, Better Dams,” ICOLD-CIGB 2025, Chengdu, China.