Artificial Intelligence for Product Quality Inspection in Manufacturing Industry - Online Detection of Edge Defects on Inorganic Solid Material
Bibliographic record
Abstract
Abstract The detection and localization of small and tiny defects on high-resolution images is considered one of the main challenges in the field of computer vision. In the manufacturing industry, the production speed and cycle time are considered the major target of a production process. For such reason, automated quality detection is getting even more complexified by the need of performing defect detection on moving products. In this work, we investigate the performance of a small defect detection process on high-scale images by utilizing state-of-the-art object detection architectures and a set of the hardware setup. Taking as a target the detection of defects on moving products, and using a small training set and a procedure of data augmentation, we demonstrated that such a challenge can be solved using machine learning and artificial intelligence coupled with domain knowledge in machine vision hardware selection and design. The sections of this paper are organized into two parts, the first part describes the problem, the existing and related works, and a summary of the existing methodologies. The second part of the paper is centered on our case study for which we started by describing the adopted methodology, the vision system design, the data acquisition and processing, the model training, and the obtained results, then it is concluded with a discussion of the model outputs and the listing of challenges that still to be studied in future works.
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) | 0.000 | 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 | 0.000 |
| 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".