ugahost991/disease-cash-crops

🌱 Plant Disease RAG System - Dataset Card

Overview

This Plant Disease RAG (Retrieval-Augmented Generation) system uses a comprehensive multi-crop disease database built from 5 Kaggle datasets, containing ~30,000+ images across multiple crop diseases.

---

🌾 Supported Crops

The system currently supports disease detection and diagnosis for the following 5 major Cash crops:

1. � Cassava (Manihot esculenta)

• Dataset: Cassava Leaf Disease Classification
• Source: Kaggle - nirmalsankalana/cassava-leaf-disease-classification
• Disease Classes (5):
  • Bacterial Blight
  • Brown Streak Disease
  • Green Mottle
  • Mosaic Disease
  • Healthy

2. � Sugarcane (Saccharum officinarum)

• Dataset: Sugarcane Plant Diseases Dataset
• Source: Kaggle - akilesh253/sugarcane-plant-diseases-dataset
• Disease Classes (6):
  • Bacterial Blight
  • Healthy
  • Mosaic
  • Red Rot
  • Rust
  • Yellow Leaf Disease

3. ☕ Coffee (Coffea spp.)

• Dataset: Coffee Leaf Diseases
• Source: Kaggle - badasstechie/coffee-leaf-diseases
• Disease Classes (4):
  • Coffee Rust (Hemileia vastatrix)
  • Coffee Leaf Miner
  • Phoma Leaf Spot
  • Healthy

4. 🍫 Cacao/Cocoa (Theobroma cacao)

• Dataset: Cacao Diseases
• Source: Kaggle - zaldyjr/cacao-diseases
• Disease Classes (3):
  • Black Pod Rot (Phytophthora)
  • Healthy
  • Pod Borer

5. 🚬 Tobacco (Nicotiana tabacum)

• Dataset: Tobacco Leaf Disease Segmentation
• Source: Kaggle - chzili/dataset-for-tobacco-leaf-disease-segmentation
• Disease Classes: Multiple disease types

---

📊 Dataset Statistics

Crop	Number of Disease Classes	Approximate Images
Cassava	5	~5,000+
Sugarcane	6	~6,000+
Coffee	4	~4,000+
Cacao	3	~3,000+
Tobacco	Multiple	~12,000+
TOTAL	20+ classes	~30,000+ images

---

🎯 Use Cases

✅ What You CAN Do:

• Upload images of Cassava, Sugarcane, Coffee, Cacao, or Tobacco leaves
• Get disease identification and similar case matches
• Receive treatment recommendations based on similar cases
• Compare your plant images with the database of known diseases
• Get confidence scores for disease predictions

❌ What You CANNOT Do (Yet):

• Detect diseases in crops NOT listed above (e.g., tomato, potato, wheat, rice, maize)
• Identify pests or insects (system focuses on diseases)
• Analyze soil conditions or nutrient deficiencies
• Detect diseases in plant parts other than leaves (roots, stems, fruits)

---

🔬 Technical Details

Model Architecture

• Base Model: Microsoft ResNet-50
• Feature Extraction: Pre-trained CNN features
• Vector Database: FAISS (GPU-accelerated)
• Embedding Dimension: 2048
• Similarity Metric: Cosine similarity (Inner Product)

Image Processing

• Input Size: 224x224 pixels
• Format: RGB images
• Supported Formats: JPG, JPEG, PNG
• Batch Processing: GPU-optimized (batch size: 64)

Search Configuration

• Default Top-K: 5 similar images
• Similarity Threshold: 0.7
• Normalization: L2 normalized embeddings

---

📝 Dataset Structure

Each image in the database contains the following metadata:

{
  "image_path": "/path/to/image.jpg",
  "class": "cassava_bacterial_blight",
  "dataset": "cassava",
  "filename": "image.jpg"
}

Class Naming Convention

Classes are prefixed with the dataset name for clarity:

• cassava_bacterial_blight
• sugarcane_rust
• coffee_rust
• cacao_black_pod_rot
• tobacco_[disease_name]

---

🚀 Getting Started

For Users

1. Check if your crop is supported: Refer to the list above
2. Prepare your image: Clear photo of the affected leaf
3. Upload to the system: Use the web interface or API
4. Review results: Get similar cases and recommendations

For AI Engineers & Developers

AI engineers can use this RAG system as a foundation to build their own plant disease detection applications, integrate it into existing agricultural platforms, or extend it with additional crops and features.

---

🔮 Comprehensive RAG Database Available

A comprehensive Plant Disease RAG database with 14+ crops and 38+ disease classes has been provided and is available on Hugging Face:

🔗 Access the Full Database: https://huggingface.co/datasets/ugahost991/plant-disease-detector

Additional Crops Included in Full Database:

• 🍅 Tomato (10 classes) - Target Spot, Septoria Leaf Spot, Early Blight, Late Blight, Bacterial Spot, Leaf Mold, Spider Mites, Yellow Leaf Curl Virus, Mosaic Virus, Healthy
• 🌶️ Pepper/Bell Pepper (2 classes) - Bacterial Spot, Healthy
• 🌽 Corn/Maize (4 classes) - Northern Leaf Blight, Common Rust, Cercospora Leaf Spot, Healthy
• 🥔 Potato (3 classes) - Early Blight, Late Blight, Healthy
• 🍎 Apple (4 classes) - Apple Scab, Black Rot, Cedar Apple Rust, Healthy
• 🍇 Grape (4 classes) - Black Rot, Esca (Black Measles), Leaf Blight, Healthy
• 🍒 Cherry (2 classes) - Powdery Mildew, Healthy
• 🍓 Strawberry (2 classes) - Leaf Scorch, Healthy
• 🍑 Peach, 🫐 Raspberry, 🫘 Soybean, 🎃 Squash, and more

How to Use the Full Database

AI engineers and developers can:

1. Download the complete RAG database from Hugging Face
2. Integrate it into their applications
3. Build custom plant disease detection systems
4. Extend with additional crops and features

Contact for Custom Solutions

For custom crop additions or enterprise solutions:

• 📧 Email: info@gss-tec.com
• 🏢 Gaston Software Solutions LLP

---

📚 Citation & Credits

Datasets Used

1. Cassava: Nirmalsankalana (2024). Cassava Leaf Disease Classification. Kaggle.
2. Sugarcane: Akilesh253 (2024). Sugarcane Plant Diseases Dataset. Kaggle.
3. Coffee: Badasstechie (2024). Coffee Leaf Diseases. Kaggle.
4. Cacao: Zaldyjr (2024). Cacao Diseases. Kaggle.
5. Tobacco: Chzili (2024). Dataset for Tobacco Leaf Disease Segmentation. Kaggle.

Model

• ResNet-50: Microsoft Research (2015)
• FAISS: Facebook AI Research

---

⚠️ Important Notes

Limitations

• Crop-Specific: Only works with the 5 crops listed above
• Leaf-Based: Designed for leaf disease detection only
• Image Quality: Requires clear, well-lit images for best results
• Not a Replacement: Should complement, not replace, expert diagnosis

Best Practices

• Use high-resolution images (minimum 224x224 pixels)
• Ensure good lighting and focus
• Capture the affected area clearly
• Include some healthy tissue for context
• Avoid blurry or dark images

---

📞 Support & Contact

Gaston Software Solutions LLP

Providing AI and ML Applications

• 📧 Email: info@gss-tec.com
• 📍 Location: Ntinda, Kampala, Uganda
• 🌐 Services: AI and Machine Learning Solutions

Documentation & Resources

For questions about supported crops or dataset issues:

• Check the README for general information
• Review QUICKSTART.md for usage instructions
• See COLAB_SETUP.md for database building

---

📄 License & Ownership

MIT Open Source License

This Plant Disease RAG System and database are provided by Gaston Software Solutions LLP (GSS-LLP) under the MIT Open Source License.

Important Notice:

• ✅ Free to use for research, commercial, and educational purposes
• ✅ Open source - modify and distribute as needed
• ⚠️ Property of GSS-LLP - The intellectual property and ownership remain with Gaston Software Solutions LLP
• ❌ No company or individual may claim ownership of this work
• ✅ Attribution required - Must credit GSS-LLP when using this system

MIT License Terms

Copyright (c) 2026 Gaston Software Solutions LLP

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Proper Attribution

When using this system, please include:

Plant Disease RAG System
Developed by: Gaston Software Solutions LLP
License: MIT
Dataset: https://huggingface.co/datasets/ugahost991/plant-disease-detector

---

Last Updated: June 2026 Version: 2.0 Dataset Link: https://huggingface.co/datasets/ugahost991/plant-disease-detector Supported Crops: 14+ (Full database) Disease Classes: 38+ Developed & Owned by: Gaston Software Solutions LLP License: MIT Open Source