High-resolution mapping of Blueberry scorch virus incidence using RGB and multispectral UAV images and deep learning
Bibliographic record
Abstract
Blueberry scorch caused by Blueberry scorch virus (BlScV) is a destructive disease, which can result in substantial yield decline and pose a significant threat to the viability of well-established highbush blueberry fields in North America and other regions. Early detection of the disease in the field, removal of infected bushes, and control of its spread via aphids to other fields or regions are critical for managing this disease. Visual assessment of Blueberry scorch symptoms is the predominant method for identifying and estimating the disease, which, however, is labour-intensive, tedious, and inefficient. Unmanned Aerial Vehicle (UAV)-based imaging is a powerful remote sensing tool for crop monitoring with several advantages, such as flexibility to acquire images of different pixel sizes, short revisit time intervals, reduced susceptibility to cloud interference, and flexibility to equip with different sensors. This study aims to collect UAV images to detect and map BlScV-infected blueberry plants using a cutting-edge deep learning model. Images of different pixel sizes acquired by an RGB sensor, and a multispectral sensor were compared to evaluate their detection accuracies. To ensure comprehensive information dependency extraction at close-, mid-, and long-ranges, the deep learning techniques developed in this study incorporate various computer vision-based mechanisms, such as Multi-Layer Perceptron (MLP), Convolutional Neural Networks (CNNs), and Self-Attention (SA) modules. Through these innovations, the deep learning algorithm, called InceptionLSA, obtained the highest average accuracy of 76.33% and 70.00% at a 20 cm pixel size of the multispectral and RGB images, respectively.
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 | 0.001 |
| 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 | 0.000 |
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".