Image-based recognition using advanced neural networks can aid surveillance of Agrilus jewel beetles
Bibliographic record
Abstract
The genus Agrilus includes two species, Agrilus planipennis and A. anxius , that are of particular phytosanitary concern and that are regulated by the European Union legislation. This implies that phytosanitary agencies of all EU countries are obliged to establish specific surveillance programs to verify the absence of these species from their territory. These activities commonly consist of the use of green-colored traps, which are however attractive not only for A. planipennis and A. anxius, but also for a wide range of other Agrilus species. For this reason, much time and expertise is required to sort and identify specimens to species, impeding an efficient rapid response. In this study, we tested the efficacy of the Entomoscope, a low-cost, open-source photomicroscope that uses high-resolution digital imaging and allows a pre-trained Convolutional Neural Networks (CNN) model to accurately detect, image and classify insect specimens, for automatic identification of 13 Agrilus species, including A. planipennis and A. anxius . We benchmarked models from three different CNN architectures and selected YOLOv8l as the most robust performer; this model achieved a Top-1 accuracy of 90.2% on a “real-world” test set (i.e. a dataset simulating real surveillance conditions). For most species, including A. planipennis and A. anxius , either no errors or only a few errors were made, whereas for a few native species misidentifications were more common. These results provided proof of concept for an AI-driven surveillance system that can strongly aid in surveillance activities of Agrilus species.
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 | 0.000 |
| Bibliometrics | 0.001 | 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) | 0.001 | 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".