Enhancing UAV-Based Precision Agriculture: A Deep Learning Approach to Targeted Maize Disease Management
Bibliographic record
Abstract
Corn/maize is the third most globally produced crop, and there are over eighty diseases impacting it, creating significant concerns in the agricultural sector. As a result, Unmanned Aerial Vehicles (UAVs) and their ability to spray specifically diseased crops have become increasingly popular. When using UAVs, Normalized Difference Vegetation Index (NDVI) calculations are frequently used to determine whether a crop should be sprayed with pesticides. However, NDVI alone is not a sufficient metric. NDVI indicates plant stress in general, which could be drought or poor soil quality, and this will not be addressed by pesticides. The solution is then to create a custom deep learning model that uses a sufficiently low NDVI value as a starting point. Once a low NDVI value is detected, the same near infrared imagery used to retrieve the NDVI calculations will be used in the model where object detection will be used to determine whether there are diseased lesions on the maize. Near infrared imagery has been proven to increase the detection of maize diseases, achieving a higher accuracy than RGB imagery. The model will mimic aspects of YOLOv10 to facilitate rapid detection times, providing the opportunity for real-time spraying. While scholars have modified YOLOv5 to achieve detection times as low as 0.064 s/sheet, a custom deep learning model used only on the contingency of low NDVI values is less computationally intensive and should then take less time overall. Finally, Keyhole Markup Language (KML) and Geographic Information Systems (GIS) will be used to create an output map indicating where pesticides were sprayed. The research will reduce the number of false positive results, which will limit the quantity of pesticides used, consequently reducing financial costs and negative environmental impacts.
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 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 | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".