Computer Vision and Deep Learning for Precise Agriculture: A Case Study of Lemon Leaf Image Classification
Bibliographic record
Abstract
Abstract Crop protection, an crucial field of precise agriculture, requires attention and improvement, as it secures sustainability and safety of crop and food production. There are various threats to crops in which pest is one of the severest. Computer vision technologies based on deep learning have shown great advantages on image classification as they enable real-time pest recognition on devices with cameras, such as drones. Thus, it is promising for pest monitoring and control and many DL models have been developed. Furthermore, early and accurate diagnosis is need as it minimizes pest damage. However, traditional models are limited on speed because the massive parameters require huge computing resource. In this work, we investigate the capability of lightweight model based on DL for the task of leaf disease classification on uncontrolled environment and compare it with traditional DL model. Lightweight models, in general, are designed to reduce computation on convolution layers with acceptable accuracy lose. We use an open database named LeLePhid, which contains lemon leave images, healthy or affected by aphid. The damage caused by aphid is general as the pest makes obvious changes to leaf outlooks. We focus on two typical DL models: the traditional, DenseNet and the lightweight, MobileNet, and discuss the balance between speed and accuracy, in order to support real-time analytics. Finally, we discuss the challenges and opportunities in practice.
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 | 0.001 | 0.001 |
| 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 | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| 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".