| # Classification of Sugarcane Leaf Disease | |
| ## Model Description | |
| The model is based on EfficientNet architecture and has been fine-tuned on a balanced dataset containing six classes: | |
| - **Bacterial Blight Disease** | |
| - **Healthy Leaves** | |
| - **Mosaic Disease** | |
| - **Red Rot Disease** | |
| - **Rust Disease** | |
| - **Yellow Disease** | |
| The model accepts RGB images of sugarcane leaves and outputs the predicted disease class. | |
| ## Dataset | |
| The dataset used for training consists of **19926 images** of sugarcane leaves, evenly distributed across the six disease classes. Each image has been pre-processed and augmented to improve model performance. | |
| ### Data Augmentation Techniques Used: | |
| - Random rotation | |
| - Flipping | |
| - Zooming | |
| - Resizing | |
| - Cropping | |
| ## Model Evaluation | |
| Epoch [10/10], Loss: 0.2903, Accuracy: 90.28% | |
| Validation Loss: 0.3633, Accuracy: 86.32% | |
| Accuracy_test: 0.8683 | |
| ## Acknowledgements | |
| The dataset used for training can be found : https://www.kaggle.com/datasets/akilesh253/sugarcane-plant-diseases-dataset | |
| ## License : CDLA-Sharing-1.0 |