Deep learning‐based embedded license plate localisation system
Bibliographic record
Abstract
In this study the authors propose novel neural network architecture for license plate localisation (LPL) based on an inverted residual structure where the shortcut connections are between the linear bottleneck layers. This residual structure is used for feature extraction in a modified single shot detector for object detection, where standard convolutions are replaced with depthwise separable convolutions in classification layers. The proposed deep learning (DL) solution was tested against three popular international research databases and achieves state‐of‐the‐art results, proving that the proposed model is accurate and robust. Across those databases, the proposed model surpasses other recent LPL works, including DL‐based methods, in terms of accuracy and speed. The authors show the proposed architecture to have significant speedup and computational efficiency gains over other DL models, and to have fast per‐image localisation processing times sufficient for applications deployed on expensive and commodity hardware alike. Using a novel multi‐threading video capture with motion detection then inference algorithm, the authors increase computational efficiency and drop fewer frames overall, allowing for increased performance. Repeated tests show that the proposed method is well‐suited to real‐time and highly accurate LPL, regardless of hardware.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| 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) | 0.005 | 0.003 |
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".