Deep learning for enhancing wavefield image quality in fast non-contact inspections
Bibliographic record
Abstract
Ultrasonic wavefield imaging with a non-contact technology can provide detailed information about the health status of an inspected structure. However, high spatial resolution, often necessary for accurate damage quantification, typically demands a long scanning time. In this work, we investigate a novel methodology to acquire high-resolution wavefields with a reduced number of measurement points to minimize the acquisition time. Such methodology is based on the combination of compressive sensing and convolutional neural networks to recover high spatial frequency information from low-resolution images. A data set was built from 652 wavefield images acquired with a laser Doppler vibrometer describing guided ultrasonic wave propagation in eight different structures, with and without various simulated defects. Out of those 652 images, 326 cases without defect and 326 cases with defect were used as a training database for the convolutional neural network. In addition, 273 wavefield images were used as a testing database to validate the proposed methodology. For quantitative evaluation, two image quality metrics were calculated and compared to those achieved with different recovery methods or by training the convolutional neural network with non-wavefield images data set. The results demonstrate the capability of the technique for enhancing image resolution and quality, as well as similarity to the wavefield acquired on the full high-resolution grid of scan points, while reducing the number of measurement points down to 10% of the number of scan points for a full grid.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".