Overview

In this tutorial, we’ll create a facial recognition-based attendance system using Python. This project combines computer vision, machine learning, and database management to automate attendance tracking for workplaces, schools, or events. By the end, you’ll have a working application that can detect faces, recognize individuals, and log their attendance into a database.


Why Build a Facial Recognition Attendance System?

Traditional attendance systems like manual sign-ins or RFID cards are time-consuming and prone to errors. Facial recognition offers a fast, contactless, and reliable solution.


Step 1: Setting Up the Environment

Install Required Libraries

To get started, install the necessary Python libraries:


pip install opencv-python dlib face-recognition sqlite3

Step 2: Preparing the Dataset

Capture or collect images of individuals whose attendance you want to track. Organize the images in directories named after the individuals for easy labeling.

Example folder structure:


dataset/
    John/
        img1.jpg
        img2.jpg
    Sarah/
        img1.jpg
        img2.jpg

Step 3: Encoding Faces

Use the face_recognition library to encode faces from the dataset. This step creates unique numerical representations for each face.


import os
import face_recognition
import cv2
import pickle

# Path to dataset
DATASET_PATH = "dataset"
ENCODINGS_FILE = "encodings.pickle"

def encode_faces():
    known_encodings = []
    known_names = []

    # Iterate through each person's folder
    for person in os.listdir(DATASET_PATH):
        person_path = os.path.join(DATASET_PATH, person)
        if not os.path.isdir(person_path):
            continue

        # Process each image
        for img_file in os.listdir(person_path):
            img_path = os.path.join(person_path, img_file)
            image = face_recognition.load_image_file(img_path)
            face_encodings = face_recognition.face_encodings(image)

            if face_encodings:
                known_encodings.append(face_encodings[0])
                known_names.append(person)

    # Save encodings to a file
    with open(ENCODINGS_FILE, "wb") as f:
        pickle.dump({"encodings": known_encodings, "names": known_names}, f)

    print("Encodings saved!")

encode_faces()

Step 4: Building the Attendance System

Load Encodings and Start the Webcam

Use the saved encodings to identify individuals in real-time.


import cv2
import face_recognition
import pickle
from datetime import datetime

# Load encodings
with open("encodings.pickle", "rb") as f:
    data = pickle.load(f)

# Initialize webcam
video_capture = cv2.VideoCapture(0)

# Track attendance
attendance_log = set()

while True:
    ret, frame = video_capture.read()
    if not ret:
        break

    # Resize frame for faster processing
    small_frame = cv2.resize(frame, (0, 0), fx=0.25, fy=0.25)
    rgb_frame = small_frame[:, :, ::-1]

    # Detect faces and compare
    face_locations = face_recognition.face_locations(rgb_frame)
    face_encodings = face_recognition.face_encodings(rgb_frame, face_locations)

    for face_encoding, face_location in zip(face_encodings, face_locations):
        matches = face_recognition.compare_faces(data["encodings"], face_encoding)
        name = "Unknown"

        if True in matches:
            match_index = matches.index(True)
            name = data["names"][match_index]

            # Log attendance
            if name not in attendance_log:
                attendance_log.add(name)
                print(f"{name} marked present at {datetime.now()}")

        # Display bounding box and name
        top, right, bottom, left = [v * 4 for v in face_location]
        cv2.rectangle(frame, (left, top), (right, bottom), (0, 255, 0), 2)
        cv2.putText(frame, name, (left, top - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)

    # Show video feed
    cv2.imshow("Attendance System", frame)

    if cv2.waitKey(1) & 0xFF == ord("q"):
        break

video_capture.release()
cv2.destroyAllWindows()

Step 5: Logging Attendance to a Database

Store the attendance data in an SQLite database for record-keeping.


import sqlite3

# Connect to SQLite database
conn = sqlite3.connect("attendance.db")
cursor = conn.cursor()

# Create table
cursor.execute("""
    CREATE TABLE IF NOT EXISTS attendance (
        id INTEGER PRIMARY KEY AUTOINCREMENT,
        name TEXT,
        timestamp DATETIME DEFAULT CURRENT_TIMESTAMP
    )
""")

def log_attendance(name):
    cursor.execute("INSERT INTO attendance (name) VALUES (?)", (name,))
    conn.commit()

# Update the webcam loop to log attendance
if name not in attendance_log:
    attendance_log.add(name)
    log_attendance(name)
    print(f"{name} marked present in the database!")

Step 6: Scheduling the System

Use schedule to automate the script to run at specific times (e.g., every morning):


import schedule
import time

def start_system():
    # Call your main attendance function here
    pass

schedule.every().day.at("08:00").do(start_system)

while True:
    schedule.run_pending()
    time.sleep(1)

Advanced Features

  1. Email Notifications: Send attendance summaries via email using smtplib.
  2. Cloud Integration: Sync attendance records with Google Sheets or a cloud database.
  3. Face Mask Detection: Add a feature to detect whether individuals are wearing masks.
  4. Multi-Camera Support: Extend the system to support multiple cameras for large setups.