Efficient Method Using Attention Based Convolutional Neural Networks for Ceramic Tiles Defect Classification
Bibliographic record
Abstract
One of the main ways of economic development has been to improve the performance and efficiency of manufacturing systems. Indeed, competition is fierce between most industrial organizations to provide new products and richer services for customers who are increasingly demanding. Although, indeed, most of the quality control and visual inspection tasks are performed by humans who use only the naked eye to detect defects, this form of control presents a number of limits, such as the size of the defects impossible to detect, risk of marketing a defective product and reduce the production performance. In this paper, a new method based on deep learning in the context of ceramic tile defect detection on a conveyor is proposed in order to provide effective quality control and real-time inspection. Our model is based on a convolutional architecture with a convolutional block attention module (CBAM) to pay more attention to the relevant areas of the input image and overcome the spatial information loss problems. A pre-processing step is performed before training by processing each image corresponding to a type of defect with an appropriate mask to facilitate learning. The experimental results show that our model produces an accurate and efficient classification of ceramic tile defects with a reduced number of parameters. We also propose a novel Ceramic defect tile dataset obtained from a ceramic production unit. The results of the experiments show that the suggested approach reaches an average accuracy rate of 99.93% compared to the state-of-the-art.
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) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".