Machine learning on thermographic images for the detection and classification of damage on composites
Bibliographic record
Abstract
Composite materials unarguably represent the important structure parts in most modern transport applications such as the aerospace sector. One area that shows great potential in the battle against aircraft structural damage and the diagnosis of composite materials. Very often, detection and diagnosis tools offer a valuable and quick mechanism to the analysts and assist them in the monitoring of the health integrity of the composite materials. Although numerous initiatives to develop damage detection techniques and make operations more efficient were launched, there is still an on-going need to develop/improve upon the existing methods. In this work, Pulsed Thermography (PT) technique was used to acquire healthy and faulty datasets from specially designed composite samples of the same dimensions (300 mm x 300 mm x 2 mm) with three different geometries (planar, curved and trapezoidal). Three plates from carbon fibre-reinforced plastic (CFRP) were tested. The same defects distribution was first introduced to the different samples and the variation of surface temperature over time, and the flow of transient heat generated through an energy stimulus in the samples were then monitored. A machine learning (A Cubic Spine Support Vector Machine) based technique was applied to the resulting thermographic images in order to detect and classify damage on composite structures. The proposed classification model was evaluated for its performance using the common metrics such as the overall accuracy, sensitivity, precision, specificity, etc. It was concluded that the classification approach could provide a reliable estimate of composite material conditions and eventually could lead to 'go / no-go' decisions.
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) | 0.000 | 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 | 0.000 |
| Open science | 0.000 | 0.000 |
| 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".