🌾 PlantScan - AI-Powered Agricultural Assistant

A smart, farmer-first platform that leverages AI for plant disease detection, government scheme recommendations, and mandi price optimization, empowering farmers to make data-driven decisions. Demo-Video of Setup-https://drive.google.com/file/d/1azFaCLcyD21dVNXEoDwOuoaJ45GE45_V/view?usp=sharing Demo-Video of Working-https://drive.google.com/file/d/14hEQFI75eVP4dKt68AtnQMLFpxTGms8M/view?usp=sharing

🎯 Problem Statement

Farmers face challenges in identifying crop diseases early, accessing relevant government schemes, and maximizing crop profits due to fragmented information and lack of technical tools.

PlantScan bridges this gap by providing a unified AI-powered platform with image analysis, localized scheme suggestions, and mandi market analysis.

🚀 Features

• 📸 Plant Disease Detection: Upload images to instantly identify crop diseases and receive treatment suggestions
• 🏛️ Scheme Recommendation: Get relevant government schemes tailored to your state, district, and needs
• 📊 Mandi Price Advisor: Discover the best locations to sell crops with smart price comparison
• 🔐 User Authentication: Secure and streamlined login/registration system
• 💻 Modern UI: Sleek, responsive interface built with React and Tailwind

🏗️ Architecture

This project follows a microservices architecture with three main components:

• Frontend: React.js application with Vite
• Backend: Node.js/Express.js API server
• LLM Service: Python FastAPI service for AI-powered features

📁 Project Structure

Mock_Hack2/
├── Frontend/                 # React.js frontend application
│   ├── src/
│   │   ├── components/       # React components
│   │   ├── data/            # Static data files
│   │   └── App.jsx          # Main application component
│   ├── package.json
│   └── vite.config.js
├── Backend/                  # Node.js backend API
│   ├── controllers/         # API controllers
│   ├── models/             # Database models
│   ├── routes/             # API routes
│   ├── middleware/         # Authentication middleware
│   └── index.js            # Server entry point
└── llm/                    # Python AI service
    ├── app.py              # FastAPI application
    ├── utils.py            # AI utility functions
    ├── schemes.py          # Government schemes data
    └── consts.py           # Constants and configurations

🛠️ Prerequisites

Before running this project, make sure you have the following installed:

• Node.js (v18 or higher)
• Python (v3.8 or higher)
• MongoDB (running locally or cloud instance)
• Git

🔧 Setup Instructions

1. Clone the Repository

git clone <repository-url>
cd Mock_Hack2

2. Environment Variables Setup

Create .env files in the following directories:

Backend/.env

MONGODB_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret_key

llm/.env

GROQ_API_KEY=your_groq_api_key
GEMINI_API=your_gemini_api_key
GOV_API=your_government_api_key

3. Backend Setup

cd Backend
npm install
nodemon index.js

The backend server will run on http://localhost:3000

4. LLM Service Setup

cd llm
pip install -r requirements.txt
uvicorn app:app --reload

The LLM service will run on http://localhost:8000

5. Frontend Setup

cd Frontend
npm install
npm run dev

The frontend application will run on http://localhost:5173

🔑 API Keys Required

To use all features, you'll need the following API keys:

1. Groq API Key: For LLM-powered responses

2. Google Gemini API Key: For image analysis

3. Government Data API Key: For mandi price data

   • Visit data.gov.in
   • Register and get your API key

🚀 Running the Application

1. Start MongoDB
2. Start the Backend: cd Backend && nodemon index.js
3. Start the LLM Service: cd llm fastapi dev app.py
4. Start the Frontend: cd Frontend && npm run dev

Open your browser and navigate to http://localhost:5173

📱 Available Routes

Frontend Routes

• / - Authentication page
• /home - Main dashboard
• /schemes - Government schemes finder
• /scan - Plant disease detection
• /marketplace - Mandi price analysis

Backend API Endpoints

• POST /api/auth/register - User registration
• POST /api/auth/login - User login
• GET /api/auth/logout - User logout
• POST /api/prompt/* - Various prompt-based endpoints

LLM Service Endpoints

• POST /mandi_price - Get mandi price recommendations
• POST /image_query - Analyze plant images for diseases
• POST /govt_scheme - Find relevant government schemes

---