Aero-engine damage detection and location method based on the CerberusDet-LS multi-task model
Bibliographic record
Abstract
Intelligent detection of damage types and localization of damages in aero-engines can improve the problems of large workload and low detection efficiency in aero-engine borescope inspection, providing key technical support for ensuring aviation safety and improvement of operation and maintenance efficiency. Aiming at the problems of low efficiency and poor detection effect of the existing single-task models in handling multiple tasks respectively, an improved aero-engine damage detection and location method CerberusDet-LS based on the CerberusDet multi-task model is proposed. The LSKA module was added to the SPPF architecture of the original model to enhance the model’s recognition ability on multi-scale targets. The Spatially Enhanced Attention (SEAM) Module is integrated into the design of the detection head. The original structure of the detection head is modified to Detect_SEAM to achieve the effective detection of occluded objects. Finally, it is verified on the aero-engine damage dataset and the location dataset. The results show that, compared with the original model, the improved CerberusDet-LS model has increased the mean Average Precision (mAP) of the damage dataset and the location dataset by 2% and 0.3% respectively, and the recall rates (R) has increased by 1.8% and 1.4% respectively. It indicates that this algorithm has higher detection accuracy and stronger generalization ability, promoting the intelligent level of damage detection and location of aero-engines.
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) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 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".