Deep Learning based Defect Detection Algorithm for Solar Panels
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Defect detection of solar panels plays an essential role in guaranteeing product quality within automated production lines. However, traditional manual inspection of solar panel defects suffers from low efficiency. This paper proposes an enhanced YOLOv5 algorithm (EL-YOLOv5) fused with the CBAM hybrid attention module to ensure product quality. The algorithm focuses on detecting five common types of defects that frequently appear on photovoltaic production lines, namely hidden cracks, scratches, broken grids, black spots, and short circuits. This study utilizes publicly available solar panel datasets, as well as datasets collected from actual photovoltaic production lines. These datasets are annotated accordingly and used to train the proposed algorithm. The experimental results demonstrate that the proposed algorithm achieves good performance on both the public and actual solar panel defect datasets. Particularly in actual datasets, where defect features are often less apparent and defects are smaller in size, the proposed algorithm can still detect even minor black spots.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it