This comprehensive dataset contains 123,588 images across 94 disease classes from 13 different crops commonly cultivated in the Bangladeshi agricultural region. The dataset is specifically curated to address plant disease identification challenges in South Asian agriculture, with a focus on crops that are economically and nutritionally significant to Bangladesh and the broader region.

The dataset combines multiple high-quality sources to create a unified, large-scale resource for computer vision research in agricultural disease detection, particularly tailored for Bangladeshi and South Asian farming contexts.

🌍 Regional Context

This dataset is particularly relevant for Bangladesh and surrounding South Asian regions because:

Climate Compatibility: All included crops thrive in Bangladesh's tropical and subtropical climate
Economic Importance: These crops represent major agricultural products in Bangladesh's economy
Food Security: Many of these crops are staples in Bangladeshi cuisine and nutrition
Agricultural Challenges: The diseases included are commonly encountered by Bangladeshi farmers
Research Gap: Addresses the need for region-specific agricultural AI solutions

📊 Dataset Statistics

Split	Images	Percentage
Train	86,467	70%
Valid	24,698	20%
Test	12,423	10%
Total	123,588	100%

🌱 Crops and Classes (94 Total)

🍌 Banana (9 classes)

Banana_Black_Pitting_or_Banana_Rust
Banana_Crown_Rot
Banana_Healthy
Banana_fungal_disease
Banana_leaf_Banana_Scab_Moth
Banana_leaf_Black_Sigatoka
Banana_leaf_Healthy
Banana_leaf__Black_Leaf_Streak
Banana_leaf__Panama_Disease

🥬 Cauliflower (4 classes)

Cauliflower_Bacterial_spot_rot
Cauliflower_Black_Rot
Cauliflower_Downy_Mildew
Cauliflower_Healthy

🌽 Corn (4 classes)

Corn_Blight
Corn_Common_Rust
Corn_Gray_Leaf_Spot
Corn_Healthy

🌿 Cotton (4 classes)

Cotton_Aphids
Cotton_Army worm
Cotton_Bacterial blight
Cotton_Healthy

🥭 Guava (9 classes)

Guava_fruit_Anthracnose
Guava_fruit_Healthy
Guava_fruit_Scab
Guava_fruit_Styler_end_root
Guava_leaf_Anthracnose
Guava_leaf_Canker
Guava_leaf_Dot
Guava_leaf_Healthy
Guava_leaf_Rust

🌾 Jute (3 classes)

Jute_Cescospora Leaf Spot
Jute_Golden Mosaic
Jute_Healthy Leaf

🥭 Mango (8 classes)

Mango_Anthracnose
Mango_Bacterial_Canker
Mango_Cutting_Weevil
Mango_Gall_Midge
Mango_Healthy
Mango_Powdery_Mildew
Mango_Sooty_Mould
Mango_die_back

🧡 Papaya (8 classes)

Papaya_Anthracnose
Papaya_BacterialSpot
Papaya_Curl
Papaya_Healthy
Papaya_Mealybug
Papaya_Mite_disease
Papaya_Mosaic
Papaya_Ringspot

🥔 Potato (10 classes)

Potato_Black_Scurf
Potato_Blackleg
Potato_Blackspot_Bruising
Potato_Brown_Rot
Potato_Common_Scab
Potato_Dry_Rot
Potato_Healthy_Potatoes
Potato_Miscellaneous
Potato_Pink_Rot
Potato_Soft_Rot

🍚 Rice (10 classes)

Rice_Blast
Rice_Brownspot
Rice_Tungro
Rice_bacterial_leaf_blight
Rice_bacterial_leaf_streak
Rice_bacterial_panicle_blight
Rice_dead_heart
Rice_downy_mildew
Rice_hispa
Rice_normal

🎋 Sugarcane (5 classes)

Sugarcane_Healthy
Sugarcane_Mosaic
Sugarcane_RedRot
Sugarcane_Rust
Sugarcane_Yellow

🍵 Tea (8 classes)

Tea_Anthracnose
Tea_algal_leaf
Tea_bird_eye_spot
Tea_brown_blight
Tea_gray_light
Tea_healthy
Tea_red_leaf_spot
Tea_white_spot

🍅 Tomato (9 classes)

Tomato_Bacterial_Spot
Tomato_Early_Blight
Tomato_Late_Blight
Tomato_Leaf_Mold
Tomato_Septoria_Leaf_Spot
Tomato_Spider_Mites_Two-spotted_Spider_Mite
Tomato_Target_Spot
Tomato_Tomato_Yellow_Leaf_Curl_Virus
Tomato_healthy

🌾 Wheat (3 classes)

Wheat_Healthy
Wheat_septoria
Wheat_stripe_rust

🔗 Data Sources

This dataset is compiled from the following high-quality sources:

HuggingFace Datasets

Corn, Cotton, Jute, Rice, Sugarcane, Wheat: Indian Crop Diseases Dataset
Rice (Additional): Paddy Disease Classification
Tea: Tea Sickness Dataset
Tomato: Tomato Leaf Disease Images

Mendeley Data Repository

Banana: Banana Disease Dataset
Potato: Potato Disease Dataset
Cauliflower: Cauliflower Disease Dataset
Guava: Guava Disease Dataset
Mango: Mango Disease Dataset
Papaya: Papaya Disease Dataset

💻 Usage

Loading the Dataset

from datasets import load_dataset

# Load the complete dataset
dataset = load_dataset("Saon110/bd-crop-veg-plant-disease-dataset")

# Access individual splits
train_dataset = dataset["train"]
valid_dataset = dataset["valid"] 
test_dataset = dataset["test"]

# Print dataset info
print(f"Total classes: {len(train_dataset.features['label'].names)}")
print(f"Train samples: {len(train_dataset)}")
print(f"Valid samples: {len(valid_dataset)}")
print(f"Test samples: {len(test_dataset)}")

Example Usage for Model Training

import torch
from torch.utils.data import DataLoader
from torchvision import transforms
from PIL import Image

# Define transforms
transform = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], 
                       std=[0.229, 0.224, 0.225])
])

# Process dataset
def preprocess_data(examples):
    images = [transform(img.convert('RGB')) for img in examples['image']]
    return {'pixel_values': images, 'labels': examples['label']}

# Apply preprocessing
train_dataset = train_dataset.with_transform(preprocess_data)

# Create DataLoader
train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True)

# Example training loop
for batch in train_loader:
    images = torch.stack(batch['pixel_values'])
    labels = torch.tensor(batch['labels'])
    # Your training code here
    break

Class Label Access

# Get all class names
class_names = train_dataset.features['label'].names

# Print some examples
for i, example in enumerate(train_dataset.take(5)):
    image = example['image']
    label_idx = example['label']
    label_name = example['label_name']
    print(f"Sample {i}: {label_name} (class {label_idx})")

🎯 Applications

This dataset is ideal for:

Agricultural Disease Detection: Build AI models for early disease identification
Precision Agriculture: Develop tools for Bangladeshi farmers
Computer Vision Research: Multi-class image classification studies
Transfer Learning: Pre-training for region-specific agricultural models
Mobile Applications: Create farmer-friendly disease detection apps
Agricultural Extension Services: Support digital agriculture initiatives in Bangladesh

🏆 Potential Research Areas

Region-Specific Disease Patterns: Study disease prevalence in South Asian agriculture
Multi-Crop Classification: Develop unified models across different crop types
Few-Shot Learning: Address classes with limited samples (e.g., Tea diseases)
Domain Adaptation: Transfer knowledge between different crop diseases
Explainable AI: Understanding model decisions for agricultural applications

📄 Dataset Structure

bd-crop-veg-plant-disease-dataset/
├── train/
│   ├── Banana_Black_Pitting_or_Banana_Rust/
│   ├── Banana_Crown_Rot/
│   ├── ...
│   └── Wheat_stripe_rust/
├── valid/
│   ├── Banana_Black_Pitting_or_Banana_Rust/
│   ├── Banana_Crown_Rot/
│   ├── ...
│   └── Wheat_stripe_rust/
└── test/
    ├── Banana_Black_Pitting_or_Banana_Rust/
    ├── Banana_Crown_Rot/
    ├── ...
    └── Wheat_stripe_rust/

⚖️ License

This dataset is licensed under CC BY-NC-SA 4.0. For commercial use, please contact: [email protected]

📚 Citation

If you use this dataset in your research, please cite:

@dataset{saon110_bd_crop_disease_2025,
  title={Bangladesh Multi-Crop Disease Classification Dataset},
  author={Sijon Chisty Saon},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/datasets/Saon110/bd-crop-vegetable-plant-disease-dataset}
}

🤝 Contributing

Contributions to improve and expand this dataset are welcome! Please feel free to:

Report issues or inconsistencies
Suggest additional crops relevant to Bangladesh
Contribute new disease categories
Improve data quality

📞 Contact

For questions, suggestions, or collaborations related to this dataset, please create an issue in the dataset repository or feel free to mail me : [email protected]

Keywords: Bangladesh, Agriculture, Plant Disease, Computer Vision, Deep Learning, Crop Classification, South Asia, Farming, AI, Machine Learning

Downloads last month: 32

Models trained or fine-tuned on Saon110/bd-crop-vegetable-plant-disease-dataset

Saon110/bd-crop-vegetable-plant-disease-model

Image Classification • Updated Sep 19, 2025 • 1