🌿 CropCure — AI-Powered Agricultural Platform

A smart farming web platform that provides instant plant disease diagnosis, real-time crop news, personalized indoor plant guidance, and an AI agriculture chatbot.

---

🔗 Live URLs

Service	URL
Frontend	https://cropcure-frontend.onrender.com/
Backend API	https://cropcure-backend-xqmm.onrender.com/

Note: Hosted on Render free tier — first load may take ~30s if the server is cold (spins down after 15 min of inactivity).

---

About The Project

CropCure is a comprehensive, multi-featured web platform designed to serve as a complete agricultural assistant for the modern farmer. It leverages a powerful combination of Deep Learning and Large Language Models to provide tools that enhance decision-making and plant management. The platform's goal is to make expert agricultural knowledge accessible, instant, and easy to understand.

---

---

Key Features

Feature	Description
🌱 AI Disease Diagnosis	Upload a plant leaf image → CNN identifies the disease → get treatment & prevention info from our knowledge base
💬 AI Agriculture Chatbot	Ask any farming question → powered by Groq (Llama-3.3-70b) with a crop disease knowledge base
📰 Real-Time Crop News	Latest agricultural news from India, powered by Tavily search API
🌤️ Weather & Soil Data	Location-based weather, soil moisture, and temperature data via AgroMonitoring API
🌿 Indoor Plant Guide	Specify your conditions → AI generates personalized indoor growing techniques
🗺️ Crop Planner	State-wise crop recommendations based on regional climate and soil data

---

Technology Stack

Backend (Python + Flask)

• Flask — REST API framework
• PyTorch — CNN model inference for plant disease detection
• Groq API — LLM chat (Llama-3.3-70b-versatile)
• Tavily API — Real-time news search
• OpenRouter API — Indoor plant recommendations (GPT-3.5)
• Gunicorn — Production WSGI server
• Scikit-learn — Label encoding for predictions

Frontend (React + Vite)

• React 19 with React Router
• Vite — Build tool
• Recharts — Data visualization (weather charts)
• Styled-Components — Component styling
• AgroMonitoring API — Weather & soil data

---

Project Structure

CropCure/
├── backend/
│   ├── app.py                    # Flask API (5 endpoints)
│   ├── rag_pipeline.py           # Groq chat with crop knowledge base
│   ├── plant_disease_classifier.py  # PyTorch CNN model
│   ├── models/                   # ML model weights (.pth, .pkl)
│   ├── data/crop_data.json       # Crop disease knowledge base
│   ├── requirements.txt
│   ├── wsgi.py                   # Gunicorn entry point
│   └── Procfile                  # Render process definition
└── frontend/
    └── src/
        ├── App.jsx               # Main app with chatbot widget
        ├── config/api.js         # Central API URL config
        ├── pages/                # Home, Weather, News, CropPlanner, etc.
        ├── components/           # NavBar, Footer, PlantDiseaseChat
        ├── services/weatherApi.js
        └── context/LanguageContext.jsx

---

API Endpoints

Method	Endpoint	Description
GET	/	Health check
POST	/chat	AI agriculture chatbot
POST	/predict	Plant disease detection (image upload)
GET	/news	Latest crop news (Tavily)
POST	/indoor-plants/recommend	Indoor plant recommendations
POST	/test-leaf	Test leaf verification only

---

Getting Started (Local Development)

Prerequisites

• Node.js 18+ and npm
• Python 3.10+ and pip

Installation

1. Clone the repository:

   git clone https://github.com/abhinavyy/CropCure.git
   cd CropCure

2. Setup the Backend:

   cd backend
   python -m venv venv
   venv\Scripts\activate        # Windows
   # source venv/bin/activate   # Mac/Linux
   pip install -r requirements.txt

3. Setup the Frontend:

   cd frontend
   npm install

4. Configure Environment Variables:

   backend/.env

   GROQ_API_KEY=your_groq_api_key
   OPENROUTER_API_KEY=your_openrouter_api_key
   TAVILY_API_KEY=your_tavily_api_key
   PORT=10000

   frontend/.env

   VITE_API_BASE_URL=http://127.0.0.1:10000

Running the Application

Option 1: Single Command (Recommended) 🚀

# From the root directory
npm install
npm start    # Starts both backend and frontend together

Option 2: Manual (Separate Terminals)

Terminal 1 — Backend:

cd backend
python app.py
# Runs on http://127.0.0.1:10000

Terminal 2 — Frontend:

cd frontend
npm run dev
# Runs on http://localhost:5173

---

Deployment (Render)

The app is deployed as two separate services on Render:

Service	Type	Root Directory	Build Command	Start/Publish
Backend	Web Service	backend	pip install -r requirements.txt	gunicorn wsgi:app --bind 0.0.0.0:$PORT --workers 1 --timeout 120
Frontend	Static Site	frontend	npm install && npm run build	Publish: dist

Frontend env var: VITE_API_BASE_URL = https://cropcure-backend-xqmm.onrender.com

---

Acknowledgments

Developed by Abhinav Yadav

---