AgriPredict – CHAIN X

Smart India Hackathon 2025

Problem Statement ID: SIH25030

AI-Based Crop Recommendation for Farmers

Theme: Agriculture, FoodTech & Rural Development
Category: Software
Team Name: CHAIN X

---

🚀 Project Overview

AgriPredict is an AI-driven decision support system that provides personalized, hyper-local crop recommendations to Indian farmers. It integrates satellite data, weather forecasts, and machine learning to offer precise farming advice without the need for manual soil testing.

Key Capabilities

• Real-time NASA SMAP soil moisture analysis
• Live weather forecast (OpenWeatherMap)
• AI-driven Crop Recommendation (NPK + pH + Satellite Data)
• Plant Disease Detection via Image Upload
• Generative AI Chatbot for farming queries (Voice + Text)
• Market Price Trends analysis

---

✅ Key Features

Feature	Technology / Data Source
Satellite Soil Analysis	NASA SMAP L4 Data (9 km resolution)
Weather Insights	OpenWeatherMap One Call API 3.0
Crop Recommendation	Hybrid Engine (Rule-based ML + Satellite Data)
Disease Diagnosis	Computer Vision (Image Processing)
Smart Chatbot	GenAI / LLM Integration (OpenRouter)
Market Intelligence	Historical price trends & prediction
Secure Authentication	JWT & BCrypt (Node.js Middleware)

---

🛠️ Tech Stack

Layer	Technology
Frontend	React.js, Vite, PWA, CSS Modules
Backend	Node.js, Express.js, MongoDB (Mongoose)
ML Models	Node.js Logic, Mock ML Server
AI / LLM	OpenRouter API
APIs	NASA Earthdata, OpenWeatherMap

---

📁 Project Structure

AgriPredict/
├── backend/
│   ├── src/
│   │   ├── main.js
│   │   ├── gen_ai/
│   │   └── models/
│   ├── server.js
│   ├── .env
│   └── package.json
│
├── agripredict-react-app/
│   ├── src/
│   ├── vite.config.js
│   └── package.json
│
├── ml_models/
│   ├── mock_server.js
│   ├── cropLogic.js
│   └── package.json
│
└── README.md

---

🧩 How to Run the Project

This project consists of three services:

1. ML Models
2. Backend
3. Frontend

---

1️⃣ Prerequisites

• Node.js & npm
• MongoDB Atlas or Local MongoDB
• OpenRouter API Key

---

2️⃣ Run ML Models (Mock)

cd AgriPredict/ml_models
npm install
node mock_server.js

📍 Runs on http://localhost:6001

---

3️⃣ Setup & Run Backend

cd AgriPredict/backend
npm install

🔐 Environment Variables (.env)

Create a .env file inside backend/ and configure as follows:

PORT=5001

# =========================
# OPENROUTER (LLM)
# =========================
OPENROUTER_API_KEY=your_openrouter_key
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
OPENROUTER_MODEL=your_openrouter_model

OPENROUTER_SITE_URL=http://localhost:5173
OPENROUTER_SITE_NAME=your_openrouter_site_name

# =========================
# ML & AUDIO CONFIG
# =========================
USE_MOCK_ML=true
USE_MOCK_AUDIO=true
AUDIO_STORAGE_PATH=./data/audio

# =========================
# DATABASE (MongoDB Atlas)
# =========================
MONGO_URI=your_Mongodb_url

# =========================
# AUTH (JWT)
# =========================
JWT_SECRET=your_JWT_secret_key
JWT_EXPIRES_IN=7d

Start Backend

npm run dev

📍 Backend runs on http://localhost:5001

---

4️⃣ Setup & Run Frontend

cd AgriPredict/agripredict-react-app
npm install
npm run dev

📍 Frontend runs on http://localhost:5173

---

🎥 Demo Features

🌾 Crop Recommendation

• Fetches user location (Lat/Lon)
• Uses NASA SMAP soil moisture
• Takes NPK input
• Returns AI-based crop suggestion

🌿 Disease Detection

• Upload leaf image
• ML mock server analyzes disease
• Treatment recommendation shown

🤖 AI Chatbot

• Text + Voice-based farming queries
• Powered by OpenRouter LLM

🌦️ Weather

• Real-time location-based weather forecast

---

🏁 Conclusion

AgriPredict – CHAIN X is a scalable, AI-powered agricultural platform designed for Indian farmers, enabling data-driven decisions using satellite intelligence and generative AI.

---

📌 Smart India Hackathon 2025

Theme: Agriculture, FoodTech & Rural Development