Smart Attendance System (Face Recognition + PyQt6)

📌 Overview

A desktop-based Smart Attendance System that automates employee attendance using face recognition.
Built with Python, PyQt6, OpenCV, dlib/face_recognition, PostgreSQL, and MongoDB.

This project demonstrates skills in GUI development, computer vision, databases, and threading.

---

🚀 Features

• Employee registration with photo & face encoding (stored in PostgreSQL).
• Real-time attendance marking using webcam + face recognition.
• Dual-database approach:
  • PostgreSQL → structured employee metadata & encodings.
  • MongoDB → attendance logs with timestamps.
• Duplicate prevention → ensures one log per employee per day.
• Green bounding box around detected faces for visual feedback.
• Non-blocking camera using PyQt QThread worker (avoids GUI freeze).

---

🛠️ Tech Stack

• Python 3.10+
• PyQt6 (GUI)
• OpenCV (camera handling)
• face_recognition/dlib (face encodings & matching)
• PostgreSQL (employee data)
• MongoDB (attendance logs)

---

📂 Project Structure

smart_attendance/
│   config.example.py
│   main.py
│
├───database
│       mongo_db.py
│       postgres_db.py
│
├───gui
│       attendance_window.py
│       logs_window.py
│       main_window.py
│       register_window.py
│
├───services
│       attendance_worker.py
│       face_recognizer.py
│       photo_storage.py
│
└───utils
        utils.py

---

⚙️ Setup Instructions

1️⃣ Clone the Repository

git clone https://github.com/TejasNN/smart-attendance-system.git
cd smart-attendance-system

2️⃣ Create a Virtual Environment

python -m venv venv
source venv/bin/activate  # On Linux/Mac
venv\Scripts\activate     # On Windows

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Setup Databases

PostgreSQL: Create DB & run create_tables() in postgres_db.py.

MongoDB: Ensure service is running (mongod).

5️⃣ Run the app

python main.py

---

🌟 Future Enhancements

• 📊 Export logs as Excel/PDF reports
• 🔑 Add login roles (Admin/User)
• 🛡️ Strengthen validation & security
• 📈 Build interactive dashboards for admins

---

🙌 Challenges & Learnings

• Learned PyQt threading to prevent GUI freezing.

• Designed a dual-database architecture.

• Improved face recognition accuracy with encoding optimizations.

• Implemented validation checks to avoid duplicate logs.

---

📸 Demo

---

👨🏻‍💻 Author

Developed by @TejasNN (Tejas Nagvekar)

🚀 Passionate about building scalable desktop and web applications with Python.