Conditional Pavement Crack Data Generation for Selective Data Augmentation Using GANs
Bibliographic record
Abstract
The adoption of Deep Learning techniques for automated crack detection is hindered by the lack of high-quality, balanced training datasets, a common issue resulting in detection bias and overfitting in model training. Data imbalances, where certain crack types predominate, cause detection bias and overfitting. This study addresses these challenges by implementing Conditional Wasserstein Generative Adversarial Networks (C-WGAN-GP) to selectively generate high-quality and diverse synthetic pavement crack images that augment underrepresented crack types. Images from Crack500 and CrackForest datasets were annotated and categorized into transverse, longitudinal, and alligator cracks. The contributions of the paper are (1) implementing an augmentation solution that selectively generates diverse and realistic pavement crack images, (2) solving mode collapse and training instability, common issues in training GAN models (3) and demonstrating that training with the GAN-augmented dataset improved classification metrics by 5% and achieved an overall accuracy of 83.33%. This approach not only mitigates data imbalance in the field, but also enhances the accuracy and capabilities of AI-driven pavement maintenance solutions
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".