Attendence.

py:

from sklearn.neighbors import KNeighborsClassifier

import cv2

import pickle

import numpy as np

import os

import csv

import time

from datetime import datetime

from win32com.client import Dispatch

# Speech function for notifications

def speak(message):

try:

voice = Dispatch("SAPI.SpVoice")

voice.Speak(message)

except Exception as e:

print(f"Error in speech synthesis: {e}")

# Create Attendance directory if it doesn't exist

if not os.path.exists('Attendance'):

os.makedirs('Attendance')

# Initialize video capture

video = cv2.VideoCapture(0)

if not video.isOpened():

raise Exception("Error: Could not access the webcam.")

# Load the trained face detection model

cascade_path = r'C:\Users\MONISHA D\Desktop\face+attendance\face

raise FileNotFoundError(f"Haarcascade file not found at {cascade_path}")

facedetect = cv2.CascadeClassifier(cascade_path)

# Load saved data for names and faces

try:

with open('data/names.pkl', 'rb') as w:

LABELS = pickle.load(w)

with open('data/faces_data.pkl', 'rb') as f:

FACES = pickle.load(f)

except FileNotFoundError as e:

raise FileNotFoundError("Data files not found. Ensure 'names.pkl' and 'faces_data.pkl' exist in
the 'data' directory.")

print('Shape of Faces matrix --> ', FACES.shape)

# Check consistency between FACES and LABELS

print(f"Number of faces: {FACES.shape[0]}")

print(f"Number of labels: {len(LABELS)}")

if FACES.shape[0] != len(LABELS):

raise ValueError("Mismatch between the number of faces and labels. Ensure the dataset is
consistent.")

# Initialize the KNN model and fit it with the data

knn = KNeighborsClassifier(n_neighbors=5)

knn.fit(FACES, LABELS)

# Optional: Load a background image for the UI (if exists)

imgBackground = cv2.imread("background.png") if os.path.exists("background.png") else None

COL_NAMES = ['NAME', 'TIME']

# Capture each frame from the webcam

ret, frame = video.read()

if not ret:

print("Error: Could not read frame from the webcam.")

break

gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

# Detect faces in the frame

faces = facedetect.detectMultiScale(gray, 1.3, 5)

for (x, y, w, h) in faces:

# Crop and resize the face

crop_img = frame[y:y + h, x:x + w]

resized_img = cv2.resize(crop_img, (50, 50)).flatten().reshape(1, -1)

# Predict the name using KNN

try:

output = knn.predict(resized_img)

except Exception as e:

print(f"Error during prediction: {e}")

continue

# Get timestamp for the current entry

ts = time.time()

date = datetime.fromtimestamp(ts).strftime("%d-%m-%Y")

timestamp = datetime.fromtimestamp(ts).strftime("%H:%M:%S")

# Check if the attendance file for the day already exists

exist = os.path.isfile(attendance_path)

# Draw rectangle around the face

cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 0, 255), 2)

cv2.rectangle(frame, (x, y - 40), (x + w, y), (50, 50, 255), -1)

cv2.putText(frame, str(output[0]), (x, y - 10), cv2.FONT_HERSHEY_COMPLEX, 1, (255, 255,

# Prepare attendance data

attendance = [str(output[0]), str(timestamp)]

# Save the attendance when 'o' is pressed

if cv2.waitKey(1) & 0xFF == ord('o'):

speak("Attendance Taken.")

time.sleep(2) # Short delay to avoid duplicate entries

# Write to the CSV file

with open(attendance_path, "a", newline="") as csvfile:

writer = csv.writer(csvfile)

if not exist:

writer.writerow(COL_NAMES) # Write column headers for a new file

writer.writerow(attendance)

# Overlay the frame onto the background image if it exists

if imgBackground is not None:

imgBackground[162:162 + 480, 55:55 + 640] = frame

cv2.imshow("Frame", imgBackground)

else:

cv2.imshow("Frame", frame)
 # Break the loop if 'q' key is pressed

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

break

# Release the video capture and close all windows

video.release()

cv2.destroyAllWindows()

Database.py:

import cv2

import pickle

import numpy as np

import os

# Initialize video capture

video = cv2.VideoCapture(0)

# Load the face detection classifier

facedetect = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')

# Initialize an empty list to store face data

faces_data = []

# Counter for capturing frames

i=0

# Get the user's name

name = input("Enter Your Name: ")

# Ensure the data directory exists

if not os.path.exists('data'):
 os.makedirs('data')

while True:

# Capture frame-by-frame

ret, frame = video.read()

# Convert frame to grayscale for face detection

gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

# Detect faces

faces = facedetect.detectMultiScale(gray, 1.3, 5)

for (x, y, w, h) in faces:

# Crop and resize the face

crop_img = frame[y:y+h, x:x+w]

resized_img = cv2.resize(crop_img, (50, 50)) # Resize to (50, 50)

# Collect the face data (capture every 10th frame)

if len(faces_data) < 50 and i % 10 == 0:

faces_data.append(resized_img)

# Update the counter

i += 1

# Display the number of images captured

cv2.putText(frame, str(len(faces_data)), (50, 50), cv2.FONT_HERSHEY_COMPLEX, 1, (50, 50,

# Draw a rectangle around the detected face

cv2.rectangle(frame, (x, y), (x + w, y + h), (50, 50, 255), 1)

cv2.imshow("Frame", frame)

# Stop if 50 faces are collected

if len(faces_data) == 50:

break

# Break the loop on 'q' key press

k = cv2.waitKey(1)

if k == ord('q'):

break

# Release the video capture object and close the window

video.release()

cv2.destroyAllWindows()

# Convert faces_data to numpy array and reshape for storage

faces_data = np.asarray(faces_data)

faces_data = faces_data.reshape(len(faces_data), -1) # Flatten the faces to 3750 dimensions

# Check if the names.pkl file exists, otherwise create it

if 'names.pkl' not in os.listdir('data/'):

names = [name] * 50 # Assign the name to all captured faces

with open('data/names.pkl', 'wb') as f:

pickle.dump(names, f)

else:

# If names.pkl exists, append new names

with open('data/names.pkl', 'rb') as f:

names = pickle.load(f)

names += [name] * 50 # Append the new name 50 times

with open('data/names.pkl', 'wb') as f:

# Check if the faces_data.pkl file exists, otherwise create it

if 'faces_data.pkl' not in os.listdir('data/'):

with open('data/faces_data.pkl', 'wb') as f:

pickle.dump(faces_data, f)

else:

# If faces_data.pkl exists, append new face data

with open('data/faces_data.pkl', 'rb') as f:

faces = pickle.load(f)

faces = np.append(faces, faces_data, axis=0) # Append new faces data, ensuring consistency

with open('data/faces_data.pkl', 'wb') as f:

pickle.dump(faces, f)

print("Data collection completed and saved!")