Swish-ResNet Method for Faulty Weld Detection
Bibliographic record
Abstract
Welding is a fundamental process commonly used in construction and manufacturing industries to effectively join different objects or materials together. There is a growing demand for accurate and dependable weld classification methods due to the increasing complexity and variety of welding applications. Various computer vision and machine learning (ML) based methods have been applied to identify issues with faulty welds. Deep learning (DL) methods typically outperform conventional ML approaches but require more data to train effectively and can overfit when data is limited. In this work, we propose an end-to-end DL method called “swish-ResNet” for reliable classification of good and faulty welds using limited data. We used data augmentation techniques to increase sample diversity and quantity. Furthermore, we employed swish activation in the ResNet-50 model to address issues related to saturation and overfitting. Specifically, we enhanced the ResNet-50 model by incorporating swish activation to improve its ability to detect complex patterns. Additionally, we added four additional dense layers at the end to refine keypoint selection in our method. Experiments conducted on two diverse datasets demonstrate the effectiveness of our swish-ResNet model for the reliable detection of faulty welds.
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) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".