aiqbench analysis
This research demonstrates that for CNN-based masonry crack detection, a training mix of 80% synthetic data and 20% real-world images can outperform models trained exclusively on real data. By leveraging simple overlay tools to generate synthetic defects, practitioners can significantly reduce the time and cost associated with collecting large-scale datasets for complex structural surfaces. The study validates synthetic data augmentation as a viable strategy to enhance model accuracy and generalization in infrastructure monitoring applications.
Balancing Real and Synthetic Data for CNN-based Masonry Crack Detection
arXiv:2606.08033v1 Announce Type: cross Abstract: Cracks are a critical indicator of building health, and early stage identification is fundamental to prevent harmful damages. Advances in deep learning (DL), particularly convolutional neural networks (CNNs), have enabled scalable solutions for automated crack detection. However, CNN performance strongly depends on the availability of large and diverse datasets, which is particularly challenging for complex surfaces such as masonry. Collecting sufficient real data is time-consuming, while publicly available datasets may not be adequate. To address this limitation, we explored generating synthetic crack data, which complements real data and improves training effectiveness. The real dataset consists of masonry crack images collected from buildings in Bologna and surrounding areas. In contrast, the synthetic dataset was generated using a crack overlay tool that adds cracks to background images in a controlled orientation and placement. The real dataset was used to train several DL architectures, to identify the best-performing model (InceptionV4) employed for experiments with generated data. Six training scenarios were tested in InceptionV4 by varying the ratio of real and synthetic data, with evaluation performed on a test set composed of real images using the F1-score and mean Intersection over Union (mIoU) metrics. Results show that training on synthetic data plus a modest addition of 20% real data achieves results comparable to training on real data only. Moreover, the 20/80 scenario (synthetic/real) achieved an 76% F1-score and 80% mean IoU, outperforming the real-only case. As can be seen, the method demonstrates the potential of synthetic data to reduce collection efforts while enhancing crack detection accuracy.
Abstract:Cracks are a critical indicator of building health, and early stage identification is fundamental to prevent harmful damages. Advances in deep learning (DL), particularly convolutional neural networks (CNNs), have enabled scalable solutions for automated crack detection. However, CNN performance strongly depends on the availability of large and diverse datasets, which is particularly challenging for complex surfaces such as masonry. Collecting sufficient real data is time-consuming, while publicly available datasets may not be adequate. To address this limitation, we explored generating synthetic crack data, which complements real data and improves training effectiveness. The real dataset consists of masonry crack images collected from buildings in Bologna and surrounding areas. In contrast, the synthetic dataset was generated using a crack overlay tool that adds cracks to background images in a controlled orientation and placement. The real dataset was used to train several DL architectures, to identify the best-performing model (InceptionV4) employed for experiments with generated data. Six training scenarios were tested in InceptionV4 by varying the ratio of real and synthetic data, with evaluation performed on a test set composed of real images using the F1-score and mean Intersection over Union (mIoU) metrics. Results show that training on synthetic data plus a modest addition of 20% real data achieves results comparable to training on real data only. Moreover, the 20/80 scenario (synthetic/real) achieved an 76% F1-score and 80% mean IoU, outperforming the real-only case. As can be seen, the method demonstrates the potential of synthetic data to reduce collection efforts while enhancing crack detection accuracy.
Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
arXiv:2606.08033 [cs.CV]
(or arXiv:2606.08033v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.08033 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 352 words
Related coverage
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
Handling Imbalanced Classification: What Works Better Than SMOTE
Analytics Vidhya · Jul 12, 2026
SynSur: An end-to-end generative pipeline for synthetic industrial surface defect generation and detection
arXiv · Jul 23, 2026
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · Sep 4, 2026