How manishkandari9/face-recognition-based-attendance-system works
Project type and purpose
• This is a full-stack web application built with Flask (Python) for taking and recording attendance using face recognition.
• Key purpose: authenticate users (students, teachers, admins), capture faces from live video or images, match against known faces, and record attendance in CSV logs.
File structure
• app.py: The Flask application entry point defining routes and controllers.
• instance/users.db: SQLite database storing user credentials and roles.
• static/
• face_recognition_model.pkl: Pretrained face-recognition embeddings model.
• haarcascade_frontalface_default.xml: OpenCV face detector.
• faces/: Directory of labeled face images for each enrolled user.
• templates/: Jinja2 HTML templates for login, signup, and dashboards (student, teacher, admin).
• Attendance/: Output CSV files, one per day, logging user presence.
• README.md: (empty) no architecture described here.
README for additional insights
• No content—rely solely on structure and file names.
Elements
• HTTP request/response between Browser and Flask.
• SQL queries between Flask and DB.
• Image data flow from Flask to Face Detector → Recognition Model.
• File‐write operations to CSV.
• Web components in blue,
• Data stores (DB, CSV folder) in green,
• Processing services (face detection/recognition) in orange.
Detail
• Capturing every key module (login, signup, dashboards, recognition).
• Background tasks (e.g., CSV file rotation per day).
• Where configuration files (XML, PKL) live and are loaded at startup.