Vision Transformer-Assisted Defect Classification and YOLOv8-Based Localization in Industrial Quality Control
Bibliographic record
Abstract
This paper proposes an intelligent quality control system integrating collaborative robotics and deep learning to enhance industrial defect detection. A Doosan robot equipped with a 2.5 D camera collaborates with AI models, including Vision Transformer (ViT), YOLOv8, MobileNetV2, and EfficientNet, to classify and localize product defects in real time. Experiments were conducted on a dataset of 1,050 annotated images, augmented to address class imbalance and evaluated under realistic conditions. Results show that ViT achieved 95.1 % accuracy in defect classification, while YOLOv8 reached 95.8% accuracy for defect localization, outperforming lightweight CNNs. The system enables reliable identification of issues such as missing or damaged pull tabs and supports automated sorting on production lines. These findings demonstrate the potential of combining collaborative robotics and advanced AI models to achieve efficient, accurate, and sustainable quality control in Industry 5.0..
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 | 0.000 | 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 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".