Automated concrete crack detection enhanced by deep learning and generative adversarial networks

Area of Science:

• Civil Engineering
• Computer Vision
• Materials Science

Background:

• Traditional concrete crack inspection is subjective and poses safety risks.
• Automated methods struggle with limited data and complex environments.
• Accurate crack detection is crucial for structural health monitoring.

Purpose of the Study:

• To propose an automated, high-precision method for concrete crack detection and measurement.
• To address challenges of limited annotated data and complex environments in crack detection.
• To provide a reproducible solution for structural health assessment.

Main Methods:

• Utilized a conditional generative adversarial network (cGAN) for high-fidelity crack image augmentation.
• Employed an enhanced U-Net segmentation network with channel attention and dilated residual blocks.
• Integrated a hybrid loss function (Dice coefficient + binary cross-entropy) to handle class imbalance.
• Applied fractal theory for quantifying geometric features of detected cracks.

Main Results:

• The cGAN-augmented dataset improved crack segmentation model performance, achieving 82.03% average precision (AP) and 83.24% F1 score.
• The fractal dimension-based measurement method demonstrated ≤4% error for complex crack networks (fractal dimension >1.6).
• Experimental results verified the method's stability and engineering applicability.

Conclusions:

• The proposed method offers a reproducible and automated solution for concrete crack detection and quantification.
• The integration of cGAN and enhanced U-Net significantly improves detection accuracy.
• Fractal theory provides a reliable approach for measuring complex crack geometries.