Optimized Plant Health Monitoring with CNNs and Transfer Learning for Precision Agriculture
Bibliographic record
Abstract
This study presents a deep learning-based approach for plant health monitoring, focusing on the classification of both diseases and nutrient deficiencies using the PlantVillage dataset. The dataset consists of 20,639 RGB leaf images from tomato, potato, and bell pepper plants, categorized into 15 classes representing both healthy and diseased samples. Three Convolutional Neural Network (CNN) architectures were evaluated: DenseNet121, VGG19, and a custom lightweight CNN developed specifically for efficient deployment in low-resource environments. DenseNet121, fine-tuned with the SGD optimizer, Sigmoid activation, and L2 regularization, achieved the highest validation accuracy of 98.27%. VGG19 reached 95.00% accuracy with a balanced training time, while the custom CNN achieved 92.00% accuracy, offering the advantages of faster prediction and reduced model size. To mitigate class imbalance and improve generalization, the models were trained with extensive data augmentation techniques, including random rotations, flips, and zooming. Regularization methods such as dropout and batch normalization further enhanced performance stability. The results highlight the trade-off between model accuracy and computational efficiency, emphasizing the potential of lightweight CNNs for real-time agricultural applications. Future work includes the integration of explainable AI (XAI) methods to enhance interpretability and build trust among agricultural practitioners, supporting broader adoption of deep learning systems in precision agriculture.
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 | 0.000 | 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) | 0.001 | 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".