PixTransNet: A Sensor-Aware CNN–Transformer Model for Magnetic Flux Leakage Defect Segmentation
Bibliographic record
Abstract
Magnetic Flux Leakage (MFL) is a widely used non-destructive evaluation (NDE) technique for pipeline inspection. However, its signals are highly sensitive to noise and geometric distortions, causing small defects with limited spatial coverage and subtle defects with low-contrast patterns to be embedded in noise, resulting in indistinct boundaries and irregular shapes that complicate segmentation. To address these challenges, we propose PixTransNet, a hybrid CNN–Transformer model built on a UNet encoder–decoder architecture with a ResNet18 backbone, designed to improve the segmentation and boundary localization of small and subtle defects in MFL signals. We embed pixel-aware transformer blocks into the deeper encoder stages to capture long-range dependencies and enhance the modeling of subtle and fragmented defect patterns. To further enhance the interpretation of MFL signals, we introduce a cross-attention module that selectively emphasizes signal regions with strong structural relevance, leading to more continuous and accurate defect boundaries, particularly for small defects. Extensive experiments on a large-scale dataset of 33,000 MFL images demonstrate that PixTransNet achieves notable improvements in segmentation quality, particularly in detecting small, weak, and low-contrast defects compared to existing baselines. PixTransNet achieves 48.30% IoU, representing a 1.97% improvement, and 70.73% Recall, representing a 13.06% improvement over the best-performing baseline
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".