Real-time Insulator Defect Detection and Damage Assessment on Edge for UAV-Based Inspection
Bibliographic record
Abstract
Outdoor insulators are essential for power transmission and distribution, providing both mechanical support and electrical insulation. There is a growing need for inspection and assessment of insulators as many of these insulators have reached their end of life. Outdoor insulators classical inspection methods are time consuming and labor intensive, and hence there is a growing need for more efficient techniques. Detecting defects and assessing damage automatically using drones equipped with edge devices is one of the potential solutions. However, this approach remains challenging due to limited labeled data sets and the computational constraints of edge devices. This paper addresses these challenges by introducing the Insulator Defects and Damage Assessment Dataset (IDDAD), a relabeled dataset merging CPLID and IDID datasets, which captures broken and missing discs, as well as multiple levels of insulator damage based on the number of defective discs. We perform a comprehensive evaluation of YOLO models (versions 5-12) across nano, small, and medium sizes, along with small RT-DETR v1 and v2 models, with a focus on edge deployment efficiency. Results show that YOLOv8n in NCNN format achieves the best balance of accuracy and speed, attaining a mean average precision (mAP50) of 0.942 at 11.3 frames per second (FPS).
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 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| 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 | 0.001 |
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".