Sugarcane Plant Diseases Dataset

This comprehensive dataset contains 19,926 high-resolution images of sugarcane leaves, categorized into six distinct classes representing both healthy and diseased conditions. It is designed for training machine learning models to classify and identify various diseases affecting sugarcane crops, including bacterial blight, mosaic disease, rust, red rot, yellow disease, and healthy leaves. The dataset has been enhanced with data augmentation techniques to improve model accuracy and robustness.

Dataset Breakdown:

1. Bacterial Blight Disease

   • Description: Sugarcane leaves affected by bacterial blight, characterized by water-soaked lesions that progress to necrosis.
   • Images: 4,800

2. Healthy Leaves

   • Description: Images of healthy sugarcane leaves showing no signs of disease or damage.
   • Images: 3,132

3. Mosaic Disease

   • Description: Sugarcane leaves with mosaic disease, causing patchy discoloration and mosaic patterns on the leaves.
   • Images: 2,772

4. Red Rot Disease

   • Description: Sugarcane leaves infected with red rot disease, marked by reddening of the stalk and decay, leading to visible leaf symptoms.
   • Images: 3,108

5. Rust Disease

   • Description: Leaves affected by rust disease, showing rust-like spots caused by fungal infections, typically brown or orange in color.
   • Images: 3,084

6. Yellow Disease

   • Description: Sugarcane leaves with yellowing and chlorosis, typical symptoms of yellow disease.
   • Images: 3,030

Total Images: 19,926
The dataset is well-balanced across the six categories, ensuring a diverse set of images for model training.

Data Augmentation Details:

• Random rotation (up to 90 degrees)
• Horizontal and vertical flipping
• Zooming (0.8x to 1.2x of original size)
• Resizing to a consistent dimension (e.g., 256×256)
• Central cropping (80% of the original image size)

Key Features:

• Large and diverse dataset for sugarcane disease classification
• Six disease categories, including healthy leaves and five types of diseases
• Augmented data for improved model performance
• Ideal for deep learning and machine learning applications in agriculture