Automated Impact Damage Detection Technique for Composites Based on Thermographic Image Processing and Machine Learning Classification
Bibliographic record
Abstract
Composite materials are one of the primary structural components in most current transportation applications, such as the aerospace industry. Composite material diagnostics is a promising area in the fight against structural damage in aircraft and spaceships. Detection and diagnostic technologies often provide analysts with a valuable and rapid mechanism to monitor the health and safety of composite materials. Although many attempts have been made to develop damage detection techniques and make operations more efficient, there is still a need to develop/improve existing methods. Pulsed thermography (PT) technology was used in this study to obtain healthy and defective data sets from custom-designed composite samples having similar dimensions but different thicknesses (1.6 and 3.8). Ten carbon fibre-reinforced plastic (CFRP) panels were tested. The samples were subjected to impact damage of various energy levels, ranging from 4 to 12 J. Two different methods have been applied to detect and classify the damage to the composite structures. The first applied method is the statistical analysis, where seven different statistical criteria have been calculated. The final results have proved the possibility of detecting the damaged area in most cases. However, for a more accurate detection technique, a machine learning method was applied to thermal images; specifically, the Cube Support Vector Machine (SVM) algorithm was selected. The prediction accuracy of the proposed classification models was calculated within a confusion matrix based on the dataset patterns representing the healthy and defective areas. The classification results ranged from 78.7% to 93.5%, and these promising results are paving the way to develop an automated model to efficiently evaluate the damage to composite materials based on the non-distractive testing (NDT) technique.
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) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| 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".