Mess Management System with Face Recognition

A polished full-stack mess automation system built for student access, manager approval, and face-based authentication.

---

What this project does

This repository combines:

• Face recognition login using Python + OpenCV
• React login interface with webcam capture
• Node.js bridge for image forwarding and authentication
• Mess management backend for student/manager workflows

The result is a unified system for student registration, manager approval, and secure mess access.

---

Table of Contents

• Features
• Architecture
• Tech Stack
• Repository Structure
• Getting Started
• Configuration
• Run the Services
• Database Setup
• Default Credentials
• Workflow
• Important Notes
• Troubleshooting

---

Features

• Face login via webcam capture
• Manual email/password login
• Student registration and manager approval
• Face profile storage and recognition
• Mess transactions, menu, feedback, polls, and announcements
• Email support for notifications

---

Architecture

Modules

• main.py — Python FastAPI service for face recognition
• server.js — Root Node.js bridge for face login and registration
• Mess_Management_Backend/ — Full mess backend with Express, Sequelize, and PostgreSQL
• Mess automation login page/ — React + Vite frontend for login UI

High-level flow

1. React captures webcam image
2. Node/Mess backend sends image to Python face service
3. Python compares face against stored gallery
4. Backend authenticates and returns login response

---

Tech Stack

• Backend: Node.js, Express, Sequelize
• Face AI: Python, FastAPI, OpenCV, NumPy
• Frontend: React, Vite, React Webcam
• Database: PostgreSQL
• Authentication: JWT, Bcrypt
• Mail: Nodemailer

---

Repository Structure

• /main.py — Face recognition API server
• /server.js — Root Node.js bridge
• /package.json — Root Node dependencies
• /requirments.txt — Python dependencies
• /Mess automation login page/ — React frontend
• /Mess_Management_Backend/ — Mess backend source code
• /TrainingImage/ — Training images
• /TrainingImageLabel/ — Training labels
• /dlib-19.22.99-cp310-cp310-win_amd64.whl — Windows dlib wheel

---

Getting Started

1. Start the Python face recognition service

cd Final-codes
python3 -m venv venv
source venv/bin/activate
pip install -r requirments.txt
uvicorn main:app --host 0.0.0.0 --port 8000 --reload

2. Start the Node.js face bridge

cd Final-codes1
npm install
node server.js

3. Start the React frontend

cd Final-codes1/Mess automation login page
npm install
npm run dev

4. Start the Mess management backend

cd Final-codes1/Mess_Management_Backend
npm install
npm run dev

---

Configuration

Create a .env file in Mess_Management_Backend/:

PORT=5000
DB_URL=postgres://USER:PASSWORD@HOST:PORT/DATABASE
JWT_SECRET=your_jwt_secret
EMAIL_USER=your_email@example.com
EMAIL_PASS=your_email_password
VITE_API_FACE=http://localhost:8000

---

Database Setup

Create a PostgreSQL database named facereco and use this schema for the root bridge service:

CREATE TABLE students (
  id SERIAL PRIMARY KEY,
  face_id INTEGER UNIQUE NOT NULL,
  name VARCHAR(100),
  email VARCHAR(100) UNIQUE NOT NULL,
  password VARCHAR(255) NOT NULL,
  role VARCHAR(50) NOT NULL,
  last_login TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

---

👤 Default Credentials

The seed script creates a default manager account:

email: manager@mess.com
password: abcd1234

Newly registered students must be approved by a manager before they can log in.

---

Workflow

Face recognition flow

• React captures the webcam image
• Backend sends the image to Python
• Python identifies the face from the gallery
• Backend returns login status

Mess backend flow

• Student registration
• Manager approval
• Menu and transaction tracking
• Feedback and polls
• Face profile login support

---

Important Notes

• Run Python, frontend, and backend services in separate terminals
• Ensure PostgreSQL is running and .env is configured
• Avoid port conflicts: 5000 is used by the mess backend by default
• Only one app should access the webcam at a time

---

Troubleshooting

• Face login fails if Python service is down
• Database errors indicate DB_URL or PostgreSQL access issues
• Email failures indicate incorrect EMAIL_USER/EMAIL_PASS

---

Additional Notes

• Mess automation login page/README.md has frontend-specific setup details
• The face recognition logic lives in main.py
• The mess backend code lives in Mess_Management_Backend/src/

---

Thanks for using this mess automation system!