TeaNet8: A real time Android application-based Tea Leaf Disease detection using fine-tuned transfer learning and Gradient-Weighted Class Activation Mapping visualization
Bibliographic record
Abstract
Tea is one of the most popular drinks in the world, and Bangladesh is a producer and user of it. However, diseases that impact the quality and productivity of crops can greatly impede the production of tea, impacting the final product’s quantity and quality. To prevent and control tea leaf diseases, a reliable and precise diagnosis and identification system is needed. Tea leaf infections are discovered manually, which takes time and affects crop quality and production. Detecting tea leaf disease early can lead to decreased damage to overall tea production. Advanced deep learning methods are simplifying the identification and categorization of specific illnesses in tea plants. The aim of this study is to introduce a new approach for identifying and categorizing illnesses found in tea plants by employing advanced deep learning methods. This study employs 2824 images of eight different types of leaf diseases. Preprocessing techniques such as Contrast Limited Adaptive Histogram Equalization (CLAHE), brightness adjustment, and unsharp masking were applied to enhance the dataset. Additionally, data augmentation techniques were used to increase its diversity. The proposed model shows that it can identify the type of persistent tea leaf disease with 97% accuracy.Gradient-Weighted Class Activation Mapping (Grad-CAM) visualization was employed to interpret and understand model predictions. The model demonstrated perfect accuracy for Algal Spot, Anthracnose, Gray Blight, and White Spot, with accuracy rates of 97.14% for Brown Blight, 94.59% for Healthy leaves, 94.12% for Red Spot, and 92.31% for Bird Eye Spot. Furthermore, the proposed model’s performance was compared against three pre-trained fine-tuning models. Various performance measurement indicators were used to evaluate the performance of the models utilized in the research. The results showed that the proposed model is effective in categorizing diseases in tea leaves.Finally, An Android-based system was developed employing the most effective model to aid farmers for detecting tea leaf diseases.
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) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.009 | 0.002 |
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".