DEMETER

Agriculture is still the crux of lives in India, but farmers usually stay uncertain about crop choice, fertilizer application, and soil care owing to patchy access to good information. In response to these, we offer an AI-based soil condition and crop suggestion system that was developed under the theme AI in Agriculture – Precision Farming & Resource Optimization. The solution combines soil health variables (pH, N-P-K, organic carbon, EC, soil type), weather predictions, and past yield data with real-time location inputs. In contrast to traditional advisory tools, the system specifically employs a multi-source voting system—cross-verifying AI output with government regulations, expert advice, and local wisdom—improving accuracy and farmer confidence. Developed with Python, JavaScript, PostgreSQL, and embedded APIs (TomorrowAPI(weather API), Government Agri portals), the system is designed for offline-first operation, low-end phone, and multilingual voice interfaces to access even illiterate farmers in rural locations. The system provides ranked crop suggestions along with personalized fertilizer schemes, leading to increased productivity, healthy soil sustainability, and increased farm revenues. This system boasts a unique architecture with a local offline Main agent which requests subagents in the cloud , reducing complexity and increasing offline availablity of the application. Measurement of success will be by means of adoption rates, recommendation precision, yield increases, and soil health results. This technology brings together cutting-edge AI with farmer-friendly design and seeks scalable impact across various crops, regions, and languages.

Demeter: Demeter is an offline-first, AI-powered mobile application for farmers, built with React Native (frontend) and Flask (backend). It delivers personalized agricultural insights without constant internet, with multilingual voice support.

Features: - Offline-first operation - AI-driven advisories - Local language and dialect support - Lightweight for low-cost devices

Prerequisites: General - Git - Python 3.9+ - Node.js (LTS) - npm or Yarn - Expo CLI

For React Native Development: - Android Studio (for Android emulator) or Xcode (for iOS, Mac only) - Optional: Physical device with Expo Go

Clone the Repository

```bash git clone https://github.com/Sukesh-2006-cse/demeter.git cd demeter Backend Setup (Flask) bash Copy Edit cd backend python -m venv venv

Activate virtual environment Windows: venv\Scripts\activate

macOS/Linux: source venv/bin/activate

Install dependencies pip install -r requirements.txt

Start Flask server python app.py The backend will run on:

cpp Copy Edit http://127.0.0.1:5000 Frontend Setup (React Native) bash Copy Edit cd frontend npm install

or yarn install

Start Expo development server npx expo start Run options:

Phone: Scan QR code with Expo Go

Emulator: Press a (Android) or i (iOS) in the terminal

Connecting Frontend & Backend Update the API base URL in the frontend configuration file (e.g., config.js or .env):

javascript Copy Edit export const API_URL = "http://127.0.0.1.8081:"; For testing on a real phone, replace 127.0.0.1 with your computer’s local IP address.

Stopping the App In each terminal, press:

objectivec Copy Edit CTRL + C To deactivate the Python virtual environment:

bash Copy Edit deactivate License This project is licensed under the MIT License — see the LICENSE file for details.

yaml Copy Edit