🌾 AgriSphere AI - Smart Farming Intelligence Platform

India's First AI + GIS Smart Farming Intelligence Platform

Complete seed-to-market intelligence with multi-class disease detection, digital twin mapping, yield prediction, IoT monitoring, and rural-accessible technology

🌟 Overview

AgriSphere AI is India's first comprehensive AI + GIS Smart Farming Intelligence Platform that combines cutting-edge machine learning, satellite imagery, and IoT sensors to revolutionize agriculture. From seed selection to market pricing, we provide complete farm management solutions designed specifically for Indian farmers.

🎯 Key Features

🤖 Multi-Class Disease Detection: AI analyzes leaf, stem, fruit & soil images with 95% accuracy
🌾 GIS Smart Farm Digital Twin: Complete digital twin with field boundaries, soil zones, and irrigation mapping
📊 AI Yield Prediction Engine: Predicts crop yields using weather, soil, and historical data
📡 IoT Soil Monitoring: Real-time monitoring with Firebase integration and smart irrigation
🌦️ Weather Risk Engine: AI-powered flood, drought, and heatwave alerts via SMS/WhatsApp
🎤 Voice Assistant (Hindi): Natural language commands in Hindi and regional languages
🛒 Farmer-Buyer Marketplace: Direct selling platform eliminating middlemen
⛓️ Blockchain Traceability: Supply chain tracking for premium quality assurance

📱 Platform Screenshots

Add screenshots of your dashboard here - main interface, disease detection, digital twin, yield prediction

🖼️ Main Dashboard Interface

🖼️ Disease Detection Results

🖼️ Digital Twin Mapping

🖼️ Yield Prediction Analytics 🖼️ Voice Assistant Interface

🛠️ Technology Stack

Frontend & UI

React 18.3+: Modern React with hooks and concurrent features
TypeScript 5.8+: Type-safe development
Vite: Lightning-fast build tool and dev server
Tailwind CSS: Utility-first CSS framework
shadcn/ui: Beautiful, accessible component library
Framer Motion: Smooth animations and transitions
React Router: Client-side routing
Recharts: Interactive data visualizations

AI & Machine Learning

TensorFlow.js 4.22+: Client-side machine learning
OpenAI API: Advanced AI capabilities
Custom ML Models: Plant disease detection, yield prediction
Python ML Pipeline: Scikit-learn, XGBoost, LSTM networks

Backend & Infrastructure

Firebase: Authentication, Firestore, and hosting
Flask/FastAPI: Python API server for ML inference
Mapbox: GIS mapping and digital twin
IoT Integration: Real-time sensor data processing

Development Tools

ESLint: Code linting and formatting
PostCSS: CSS processing
Vite Plugins: Optimized development experience

📊 Supported Crops & Diseases

Major Crops

Crop	Season	Disease Classes	Accuracy
🌾 Rice	Kharif	15+ diseases	96%
🌾 Wheat	Rabi	12+ diseases	95%
🌽 Maize	Kharif/Rabi	10+ diseases	94%
🥔 Potato	Rabi	8+ diseases	97%
🍅 Tomato	All seasons	20+ diseases	95%

Disease Detection Classes

Leaf Diseases: Blight, Spot, Rust, Mold
Stem Diseases: Rot, Canker, Wilt
Fruit Diseases: Rot, Spot, Blight
Soil Issues: Nutrient deficiency, pH imbalance
Pest Damage: Insect bites, fungal infections

🎯 Quick Start

Prerequisites

Node.js 18+ and npm
Python 3.8+ (for ML training)
Git
Firebase account (for authentication)

🔧 Installation

Install dependencies
```
npm install
```

2.. Install Firebase

npm install firebase

Create .env file Create a .env file in the root directory with the following content:

VITE_OPENAI_API_KEY=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
VITE_MAPBOX_ACCESS_TOKEN=pk.eyJ1IjoibXVza2FuMTIxNiIsImEiOiJjbTkzNDFoM2owYnUyMndzNDI1OG4yY3k4In0.4j6e_uHRIj9rwP8W7R658Q
VITE_WEATHER_API_KEY=796cdb2a0021887a20495ba82c2b2cc5

Start the development server
```
npm run dev
```

Access the application

Frontend: http://localhost:8080
Backend API or Python : http://localhost:5000

🐳 Python Installation

requirements_api.txt
requirements_training.txt
requirements_yield.txt
simple_requirements.txt

Start the python development server

python api_server.py

📚 Usage Guide

🎮 Disease Detection

Upload Crop Images
- Take photos of leaves, stems, fruits, or soil
- Support for JPG, PNG formats
- Real-time analysis with confidence scores

You can test the AI model using images from the sample_test/ folder included in this repository.

📤 Upload Crop Images

Take photos of leaves, stems, fruits, or soil

Or use the sample images from:

/sample_test/

Supports JPG and PNG formats

Real-time analysis with confidence scores

AI Analysis Results
- Disease identification with 95% accuracy
- Treatment recommendations
- Prevention strategies
- Cost estimates for treatments

🌾 Digital Twin Mapping

Field Boundary Drawing
- Interactive map interface
- GPS coordinate capture
- Multi-polygon support
Soil Zone Analysis
- Satellite imagery integration
- Soil type classification
- Irrigation zone mapping

📊 Yield Prediction

Input Parameters
- Weather data (temperature, rainfall, humidity)
- Soil characteristics (pH, nutrients, texture)
- Historical yield data
- Crop variety and planting date
AI Prediction Results
- Yield estimates with confidence intervals
- Risk assessment
- Optimization recommendations

🧠 Machine Learning Models

Disease Detection Model

Architecture: EfficientNetB0 + Custom Classification Head
Training Data: PlantVillage Dataset (50,000+ images)
Accuracy: 95%+ validation accuracy
Classes: 15+ disease categories
Inference: <100ms per image

Yield Prediction Models

Algorithms: Random Forest, XGBoost, LSTM
Features: Weather, soil, historical data (40+ features)
Accuracy: 92-96% depending on crop
Time Series: 7-day weather forecasting

Training Pipeline

# Disease Detection Training
1. Dataset preprocessing and augmentation
2. EfficientNetB0 base model fine-tuning
3. Custom classification head training
4. Model evaluation and export

# Yield Prediction Training
1. Feature engineering (40+ features)
2. Multi-model ensemble training
3. Cross-validation and hyperparameter tuning
4. Model serialization for production

🏗️ Project Structure

Agrisphere/
├── 📁 src/
│   ├── 📁 components/          # Reusable UI components
│   │   ├── ui/                # shadcn/ui components
│   │   ├── Login.tsx          # Authentication components
│   │   ├── Signup.tsx
│   │   └── AIChat.tsx         # AI assistant
│   ├── 📁 pages/              # Main application pages
│   │   ├── Index.tsx          # Landing page
│   │   ├── DiseaseDetection.tsx
│   │   ├── DigitalTwin.tsx
│   │   ├── YieldPrediction.tsx
│   │   ├── IoTMonitoring.tsx
│   │   ├── Marketplace.tsx
│   │   ├── VoiceAssistant.tsx
│   │   └── ComprehensiveDashboard.tsx
│   ├── 📁 store/              # State management
│   │   └── authStore.ts       # Authentication store
│   ├── 📁 lib/                # Utilities and configurations
│   │   ├── firebase.ts        # Firebase config
│   │   └── utils.ts           # Helper functions
│   └── App.tsx                # Main app component
├── 📁 public/                 # Static assets
├── 📁 models/                 # Trained ML models
├── 📁 data/                   # Training datasets
├── 📁 api_server.py          # Flask API for ML inference
├── 📁 train_*.py             # Model training scripts
├── 📁 requirements*.txt      # Python dependencies
├── 📁 package.json           # Node.js dependencies
└── 📁 README.md              # Project documentation

Environment Variables

VITE_FIREBASE_API_KEY=AIzaSyBJkpgg7K6yTyii-hBR2tCR0AX21bTQNgw
VITE_FIREBASE_AUTH_DOMAIN=agrispace-ea219.firebaseapp.com
VITE_FIREBASE_PROJECT_ID=agrispace-ea219
VITE_FIREBASE_STORAGE_BUCKET=agrispace-ea219.firebasestorage.app
VITE_FIREBASE_MESSAGING_SENDER_ID=528915442362
VITE_FIREBASE_APP_ID=1:528915442362:web:1c00c4257780e04aea3083
VITE_FIREBASE_MEASUREMENT_ID=G-WWZSNCRDH0

Firebase Hosting Alternative

# Build and deploy to Firebase
npm run build
firebase deploy --only hosting

🔧 API Documentation

Disease Detection Endpoint

POST /api/disease-detection
Content-Type: application/json

{
  "image": "base64_encoded_image",
  "crop_type": "rice",
  "location": "patiala_punjab"
}

Response:
{
  "disease": "Leaf Blight",
  "confidence": 0.96,
  "treatment": "Apply copper fungicide",
  "prevention": "Improve air circulation",
  "cost_estimate": "₹500-800 per acre"
}

Yield Prediction Endpoint

POST /api/yield-prediction
Content-Type: application/json

{
  "crop": "rice",
  "area": 2.5,
  "soil_ph": 6.8,
  "rainfall": 1200,
  "temperature": 28.5,
  "historical_yield": 4500
}

Response:
{
  "predicted_yield": 4800,
  "confidence_interval": [4200, 5400],
  "risk_level": "Low",
  "recommendations": ["Increase potassium fertilizer", "Install drip irrigation"]
}

📈 Performance Metrics

Model Accuracy

Disease Detection: 95.2% overall accuracy
Yield Prediction: 93.8% accuracy
IoT Monitoring: 99.1% uptime
Voice Recognition: 89.5% Hindi accuracy

System Performance

Frontend Load Time: <2 seconds
API Response Time: <500ms
Image Processing: <3 seconds
Concurrent Users: 1000+

🤝 Contributing

We welcome contributions from the community! Here's how you can help:

🔧 Development Setup

Fork the repository
Create feature branch
```
git checkout -b feature/amazing-feature
```

Install dependencies

npm install
pip install -r requirements_training.txt

Make your changes
Test thoroughly
Submit pull request

🐛 Bug Reports

Please use the GitHub Issues page to report bugs.

Include:

Detailed description of the bug
Steps to reproduce
Expected vs actual behavior
Screenshots if applicable
System information (OS, browser, etc.)

💡 Feature Requests

Have ideas for new features? Open an issue on GitHub with:

Clear feature description
Use case and benefits
Implementation suggestions (optional)

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

👥 Team

Muskan - Lead Developer & ML Engineer
Contributors - Welcome to join our mission!

🙏 Acknowledgments

PlantVillage Dataset for disease detection training data
OpenAI for advanced AI capabilities
Mapbox for GIS mapping services
TensorFlow.js for client-side ML
All contributors and supporters of this project

📞 Contact

For support, feature requests, or collaboration inquiries:

Made with ❤️ for Indian Farmers

Name		Name	Last commit message	Last commit date
Latest commit History 36 Commits
.ipynb_checkpoints		.ipynb_checkpoints
__pycache__		__pycache__
archive4_model_output		archive4_model_output
improved_model_output		improved_model_output
light_archive4_output		light_archive4_output
models		models
optimized_dataset/healthy		optimized_dataset/healthy
pest		pest
public		public
sample_test		sample_test
simple_archive4_output		simple_archive4_output
sklearn_model_output		sklearn_model_output
src		src
test_export_model		test_export_model
.gitignore		.gitignore
ENHANCED_FEATURES.md		ENHANCED_FEATURES.md
FIXES_SUMMARY.md		FIXES_SUMMARY.md
HACKATHON_SETUP.md		HACKATHON_SETUP.md
IMPLEMENTATION_SUMMARY.md		IMPLEMENTATION_SUMMARY.md
IMPROVEMENTS_GUIDE.md		IMPROVEMENTS_GUIDE.md
LICENSE		LICENSE
QUICK_START.md		QUICK_START.md
README.md		README.md
README_NEXT_STEPS.md		README_NEXT_STEPS.md
STATUS.md		STATUS.md
TENSORFLOW_FIX.md		TENSORFLOW_FIX.md
TEST_INSTRUCTIONS.md		TEST_INSTRUCTIONS.md
TODO.md		TODO.md
TRAINING_GUIDE.md		TRAINING_GUIDE.md
Untitled3.ipynb		Untitled3.ipynb
api_server.py		api_server.py
api_server_fixed.py		api_server_fixed.py
bun.lockb		bun.lockb
buyer_interactions.json		buyer_interactions.json
check_models.py		check_models.py
community_data.json		community_data.json
components.json		components.json
create_demo_models.py		create_demo_models.py
crop_loss_data.json		crop_loss_data.json
debug_output.txt		debug_output.txt
debug_voice.py		debug_voice.py
demands.json		demands.json
eslint.config.js		eslint.config.js
feature_importance.png		feature_importance.png
fix_numpy_tf.py		fix_numpy_tf.py
fix_numpy_version.py		fix_numpy_version.py
fix_tf_numpy.bat		fix_tf_numpy.bat
fix_tf_numpy.sh		fix_tf_numpy.sh
improved_voice_assistant.py		improved_voice_assistant.py
index.html		index.html
install_dependencies.bat		install_dependencies.bat
market_engine.py		market_engine.py
marketplace_listings.json		marketplace_listings.json
models.txt		models.txt
output.txt		output.txt
package-lock.json		package-lock.json
package.json		package.json
pest_engine.py		pest_engine.py
postcss.config.js		postcss.config.js
prediction_plots.png		prediction_plots.png
recommendation_engine.py		recommendation_engine.py
requirements_api.txt		requirements_api.txt
requirements_training.txt		requirements_training.txt
requirements_yield.txt		requirements_yield.txt
seed_gov_data.py		seed_gov_data.py
setup_environment.py		setup_environment.py
simple_requirements.txt		simple_requirements.txt
simple_train.py		simple_train.py
start_all.bat		start_all.bat
start_api.bat		start_api.bat
start_dev.bat		start_dev.bat
start_frontend.bat		start_frontend.bat
tailwind.config.ts		tailwind.config.ts
test_api.py		test_api.py
test_archive4_training.py		test_archive4_training.py
test_gemini.py		test_gemini.py
test_model_export.py		test_model_export.py
test_pest_api.py		test_pest_api.py
test_recommendation.py		test_recommendation.py
train_and_start.bat		train_and_start.bat
train_archive4.bat		train_archive4.bat
train_archive4_light.py		train_archive4_light.py
train_archive4_model.py		train_archive4_model.py
train_archive4_simple.py		train_archive4_simple.py
train_enhanced_models.py		train_enhanced_models.py
train_enhanced_plant_model.py		train_enhanced_plant_model.py
train_high_accuracy_model.py		train_high_accuracy_model.py
train_improved_model.py		train_improved_model.py
train_model.py		train_model.py
train_optimized_model.py		train_optimized_model.py
train_simple.py		train_simple.py
train_simple_demo.py		train_simple_demo.py
train_simple_improved.py		train_simple_improved.py
train_sklearn_model.py		train_sklearn_model.py
train_yield_model.py		train_yield_model.py
tsconfig.app.json		tsconfig.app.json
tsconfig.json		tsconfig.json
tsconfig.node.json		tsconfig.node.json
vite.config.ts		vite.config.ts

Folders and files

Latest commit

History

Repository files navigation