Enhancing Crack Detection on Levees with Synthetic Data Augmentation via ACGAN and Attention-Boosted Faster R-CNN


Saludin Saludin(1*); Wiwit Priyadi(2); Gita Puspa Artiani(3); Risanto Darmawan(4); Ali Khumaidi(5);

(1) Universitas Bina Insani
(2) Universitas Bina Insani
(3) Universitas Krisnadwipayana
(4) Universitas Krisnadwipayana
(5) Universitas Bina Insani
(*) Corresponding Author

  

Abstract


This study introduces an innovative approach for detecting cracks on levee surfaces by integrating an Auxiliary Classifier Generative Adversarial Network (ACGAN) for data augmentation with a Faster R-CNN model enhanced by an attention mechanism. The ACGAN-based augmentation aims to generate synthetic images that enrich data variability in the original dataset. The attention-optimized Faster R-CNN is designed to improve detection precision, particularly for small objects and fine cracks that are difficult to distinguish from the background. Experimental results demonstrate that the incorporation of ACGAN improves detection performance, increasing both the mean Average Precision (mAP) and Average Recall (AR). The model achieved an mAP of approximately 0.56 at IoU = 0.50 and 0.34 at IoU = 0.75, while the AR (maxDets = 100) reached 0.55, indicating a strong capability in identifying most crack instances. When trained on the combined dataset of original and synthetic images, the Faster R-CNN model reached a precision of 0.92 for the severe crack class, while performance for minor cracks remained lower (precision 0.78). Adjusting the confidence threshold to 0.65 improved detection reliability by reducing noise and retaining high-confidence predictions. Improved performance in detecting severe cracks supports timely maintenance and repair decisions. This study demonstrates the effectiveness of GAN-based data augmentation and attention-enhanced object detection for automated structural health monitoring (SHM) of levee infrastructure

Keywords


ACGAN; Faster R-CNN; Levee Crack Detection; Data Augmentation; Attention Mechanism

  
  

Full Text:

PDF
  

Article Metrics

Abstract view: 115 times
PDF view: 52 times 		     

Digital Object Identifier

doi  https://doi.org/10.33096/ilkom.v18i1.3206.58-68 		  

Cite

How to cite item

References


M. Farid et al., “Development of an Empirical Flood Fragility Curve for Levee Failure and Its Application in Probabilistic Flood Risk Assessment: A Case Study of Citeureup Village, Indonesia,” Int. J. Disaster Risk Reduct., p. 105793, Sep. 2025, doi: 10.1016/j.ijdrr.2025.105793.

A. Eskandarinejad, R. Nazari, M. R. Nikoo, D. Arellano, S. Pezeshk, and S. H. Ghasemi, “A comprehensive review of geotechnical implications of floods and water-driven disasters,” Sci. Total Environ., vol. 985, p. 179731, Jul. 2025, doi: 10.1016/j.scitotenv.2025.179731.

Z. Qu, Y. Zhang, Z. Liu, R. Si, and J. Wu, “A review on early-age cracking of concrete: Causes and control,” Case Stud. Constr. Mater., vol. 21, p. e03848, Dec. 2024, doi: 10.1016/j.cscm.2024.e03848.

S. J. H. Rikkert, M. Kok, K. Lendering, and R. Jongejan, “A pragmatic, performance‐based approach to levee safety assessments,” J. Flood Risk Manag., vol. 15, no. 4, Dec. 2022, doi: 10.1111/jfr3.12836.

Q. Guo, “Strategies for a resilient, sustainable, and equitable Mississippi River basin,” River, vol. 2, no. 3, pp. 336–349, Aug. 2023, doi: 10.1002/rvr2.60.

G. Yu, X. Zhou, and X. Chen, “VDCrackGAN: A Generative Adversarial Network with Transformer for Pavement Crack Data Augmentation,” Appl. Sci., vol. 14, no. 17, p. 7907, Sep. 2024, doi: 10.3390/app14177907.

T. S. Tran, S. D. Nguyen, H. J. Lee, and V. P. Tran, “Advanced crack detection and segmentation on bridge decks using deep learning,” Constr. Build. Mater., vol. 400, p. 132839, Oct. 2023, doi: 10.1016/j.conbuildmat.2023.132839.

H. Kaveh and R. Alhajj, “Recent advances in crack detection technologies for structures: a survey of 2022-2023 literature,” Front. Built Environ., vol. 10, Jul. 2024, doi: 10.3389/fbuil.2024.1321634.

G. Chen, S. Teng, M. Lin, X. Yang, and X. Sun, “Crack Detection Based on Generative Adversarial Networks and Deep Learning,” KSCE J. Civ. Eng., vol. 26, no. 4, pp. 1803–1816, Apr. 2022, doi: 10.1007/s12205-022-0518-2.

S. Lie et al., “High Quality Training Set Development for Road Crack Detection Using Progressive Data-Driven Models with Integrated Identification,” Case Stud. Constr. Mater., p. e05416, Oct. 2025, doi: 10.1016/j.cscm.2025.e05416.

L. Pei, Z. Sun, L. Xiao, W. Li, J. Sun, and H. Zhang, “Virtual generation of pavement crack images based on improved deep convolutional generative adversarial network,” Eng. Appl. Artif. Intell., vol. 104, p. 104376, Sep. 2021, doi: 10.1016/j.engappai.2021.104376.

E. Branikas, P. Murray, and G. West, “A Novel Data Augmentation Method for Improved Visual Crack Detection Using Generative Adversarial Networks,” IEEE Access, vol. 11, pp. 22051–22059, 2023, doi: 10.1109/ACCESS.2023.3251988.

D. P. Yadav, B. Sharma, S. Chauhan, and I. Ben Dhaou, “Bridging Convolutional Neural Networks and Transformers for Efficient Crack Detection in Concrete Building Structures,” Sensors, vol. 24, no. 13, p. 4257, Jun. 2024, doi: 10.3390/s24134257.

Q. Wu and R. Han, “Deep Learning Based Crack Detection Method for Bridges,” in 2025 IEEE International Conference on Electronics, Energy Systems and Power Engineering (EESPE), IEEE, Mar. 2025, pp. 449–455. doi: 10.1109/EESPE63401.2025.10987561.

Y. Sun, Z. Sun, and W. Chen, “The evolution of object detection methods,” Eng. Appl. Artif. Intell., vol. 133, p. 108458, Jul. 2024, doi: 10.1016/j.engappai.2024.108458.

C. Cao et al., “An Improved Faster R-CNN for Small Object Detection,” IEEE Access, vol. 7, pp. 106838–106846, 2019, doi: 10.1109/ACCESS.2019.2932731.

D. Dais, İ. E. Bal, E. Smyrou, and V. Sarhosis, “Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning,” Autom. Constr., vol. 125, p. 103606, May 2021, doi: 10.1016/j.autcon.2021.103606.

R. E. Philip, D. Andrushia A, N. Anand, M. E. Mathews, M. Z. Naser, and E. Lubloy, “Improved YOLOv5-based multi-crack detection in concrete wall surfaces,” Appl. Eng. Sci., vol. 23, p. 100247, Sep. 2025, doi: 10.1016/j.apples.2025.100247.

N. Zhang, X. Gao, X. Lai, H. Xu, Y. Liu, and H. Ma, “Research on crack detection method for shallow-buried underground compressed air energy storage cavern based on improved mask R-CNN model,” Earth Energy Sci., vol. 1, no. 3, pp. 203–212, Sep. 2025, doi: 10.1016/j.ees.2025.07.001.

Q. Li, X. Xu, J. Guan, and H. Yang, “The Improvement of Faster-RCNN Crack Recognition Model and Parameters Based on Attention Mechanism,” Symmetry (Basel)., vol. 16, no. 8, p. 1027, Aug. 2024, doi: 10.3390/sym16081027.

D. Matuzevičius, “A Retrospective Analysis of Automated Image Labeling for Eyewear Detection Using Zero-Shot Object Detectors,” Electronics, vol. 13, no. 23, p. 4763, Dec. 2024, doi: 10.3390/electronics13234763.

D. Ruan, X. Chen, C. Gühmann, and J. Yan, “Improvement of Generative Adversarial Network and Its Application in Bearing Fault Diagnosis: A Review,” Lubricants, vol. 11, no. 2, p. 74, Feb. 2023, doi: 10.3390/lubricants11020074.

H. Cang et al., “Jujube quality grading using a generative adversarial network with an imbalanced data set,” Biosyst. Eng., vol. 236, pp. 224–237, Dec. 2023, doi: 10.1016/j.biosystemseng.2023.11.002.

E. Thirumagal and K. Saruladha, “GAN models in natural language processing and image translation,” in Generative Adversarial Networks for Image-to-Image Translation, Elsevier, 2021, pp. 17–57. doi: 10.1016/B978-0-12-823519-5.00001-4.

X. Xu et al., “Crack Detection and Comparison Study Based on Faster R-CNN and Mask R-CNN,” Sensors, vol. 22, no. 3, p. 1215, Feb. 2022, doi: 10.3390/s22031215.

V. Sai Sandeep Velaga, “Object Detection Using Faster R CNN with ROI Alignment using Bilinear Interpolation,” Int. J. Sci. Dev. Res., vol. 10, no. 9, 2025, doi: 10.56975/ijsdr.v10i9.304835.

P. Fu and J. Wang, “Lithology Identification Based on Improved Faster R-CNN,” Minerals, vol. 14, no. 9, p. 954, Sep. 2024, doi: 10.3390/min14090954.

Q. Su, G. Zhang, S. Wu, and Y. Yin, “FI-FPN: Feature-integration feature pyramid network for object detection,” AI Commun., vol. 36, no. 3, pp. 191–203, Aug. 2023, doi: 10.3233/AIC-220183.

M. Trigka and E. Dritsas, “A Comprehensive Survey of Deep Learning Approaches in Image Processing,” Sensors, vol. 25, no. 2, p. 531, Jan. 2025, doi: 10.3390/s25020531.

A. Berger and S. Guda, “Threshold optimization for F measure of macro-averaged precision and recall,” Pattern Recognit., vol. 102, p. 107250, Jun. 2020, doi: 10.1016/j.patcog.2020.107250.

F. Liu, Y. Li, and Y. Zheng, “Time Series Data Generation Method with High Reliability Based on ACGAN,” Entropy, vol. 27, no. 2, p. 111, Jan. 2025, doi: 10.3390/e27020111.

G. Wang, Z. Yang, H. Sun, Q. Zhou, and Z. Yang, “AC-SNGAN: Multi-class data augmentation for damage detection of conveyor belt surface using improved ACGAN,” Measurement, vol. 224, p. 113814, Jan. 2024, doi: 10.1016/j.measurement.2023.113814.

Z. Sordo et al., “Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures,” J. Imaging, vol. 11, no. 8, p. 252, Jul. 2025, doi: 10.3390/jimaging11080252.

M. Ibrahim et al., “Generative AI for synthetic data across multiple medical modalities: A systematic review of recent developments and challenges,” Comput. Biol. Med., vol. 189, p. 109834, May 2025, doi: 10.1016/j.compbiomed.2025.109834.

W. A. Shobaki and M. Milanova, “A Comparative Study of YOLO, SSD, Faster R-CNN, and More for Optimized Eye-Gaze Writing,” Sci, vol. 7, no. 2, p. 47, Apr. 2025, doi: 10.3390/sci7020047.

H. Mahamivanan, J. Matthews, P. E. D. Love, and F. Nasirzadeh, “Toward accurate detection of small objects in rail construction: A deep learning perspective,” Eng. Appl. Artif. Intell., vol. 160, p. 111977, Nov. 2025, doi: 10.1016/j.engappai.2025.111977.

Q. Yuan, Y. Shi, and M. Li, “A Review of Computer Vision-Based Crack Detection Methods in Civil Infrastructure: Progress and Challenges,” Remote Sens., vol. 16, no. 16, p. 2910, Aug. 2024, doi: 10.3390/rs16162910.


Refbacks

  • There are currently no refbacks.


Copyright (c) 2026 Saludin Saludin, Wiwit Priyadi, Gita Puspa Artiani, Risanto Darmawan, Ali Khumaidi

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.