Toward smart railway maintenance: AI-enhanced Non-Destructive Evaluation using Vision Transformers and CNNs for fastener defect detection
Bibliographic record
Abstract
Predictive health management and maintenance of transport infrastructure are critical for preventing accidents and service disruptions. Applying Non-Destructive Evaluation (NDE) and imaging techniques is essential for identifying irregularities without causing harm. This research utilizes pre-trained models and incorporates transfer learning concepts to overcome dataset constraints. This study assesses the effectiveness of various machine learning models, including the Vision Transformer (ViT), Data-efficient Image Transformer (DeiT), VGG19, VGG16, and ResNet50, in enhancing NDE for railway track fasteners. ViT and DeiT, both transformer-based models, emerged as the top performers due to their superior learning efficiencies and generalization capabilities, augmented by precise hyperparameter tuning. VGG models are a reliable alternative, while ResNet50 is better suited for applications prioritizing computational efficiency over accuracy. • Transformers (ViT/DeiT) outperform CNNs in railway fastener defect detection, showing higher accuracy and robustness. • Leveraging transfer learning and pre-trained ViT/DeiT models helps overcome data scarcity in rail defect datasets. • DeiT achieved 95.04% accuracy, ViT 94.14%, both beating VGG16's 91.54% and showing lower validation loss values. • Results highlight transformers' promise in railway NDE, enabling future hybrid models and larger dataset use.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".