Skip to main content

Face Detection Based Attendance System

 Face Detection Based Attendance System

  • Create a Main Folder named "Face Detection Based Attendance System" in VS Code. 
  • Create a file named "add_faces.py"

add_faces.py

import cv2
video=cv2.VideoCapture(0)

while True:
    ret,frame=video.read()
    cv2.imshow("Frame",frame)
    k=cv2.waitKey(1)
    if k==ord('q'):
        break
video.release()
cv2.destroyAllWindows()

  • Open a new terminal and type "python add_faces.py"
  • This will open your web camera. So, the process is getting started. Click "Q" to exit camera. 

Create a Folder named "Data". In that folder, create a file named "haarcascade_frontalface_default.xml"

haarcascade_frontalface_default.xml

For, haarcascade_frontalface_default.xml code link  Click Here

Now, write code in add_faces.py as,

add_faces.py

import cv2

video=cv2.VideoCapture(0)
facedetect=cv2.CascadeClassifier('data\haarcascade_frontalface_default.xml')
while True:
    ret,frame=video.read()
    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces=facedetect.detectMultiScale(gray, 1.3 ,5)
    for (x,y,w,h) in faces:
        cv2.rectangle(frame, (x,y), (x+w, y+h), (50,50,255), 1)
    cv2.imshow("Frame",frame)
    k=cv2.waitKey(1)
    if k==ord('q'):
        break
video.release()
cv2.destroyAllWindows()

  • Open terminal and type "python add_faces.py"
  • This will open your web camera. So, the process is again started. And this time, our face is surrounded by a rectangular box. Click "Q" to exit camera. 

Now, write code in add_faces.py as,

add_faces.py

import cv2

video=cv2.VideoCapture(0)
facedetect=cv2.CascadeClassifier('data/haarcascade_frontalface_default.xml')

faces_data=[]

i=0

while True:
    ret,frame=video.read()
    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces=facedetect.detectMultiScale(gray, 1.3 ,5)
    for (x,y,w,h) in faces:
        crop_img=frame[y:y+h, x:x+w, :]
        resized_img=cv2.resize(crop_img, (50,50))
        if len(faces_data)<=100 and i%10==0:
            faces_data.append(resized_img)
        i=i+1
        cv2.putText(frame, str(len(faces_data)), (50,50), cv2.FONT_HERSHEY_COMPLEX, 1, (50,50,255), 1)
        cv2.rectangle(frame, (x,y), (x+w, y+h), (50,50,255), 1)
    cv2.imshow("Frame",frame)
    k=cv2.waitKey(1)
    if k==ord('q') or len(faces_data)==100:
        break
video.release()
cv2.destroyAllWindows()

  • Open terminal and type "python add_faces.py"
  • This will open your web camera. So, the process is again started. And this time, our face is surrounded by a rectangular box and timer will gets started. Click "Q" to exit camera. 

Now, write code in add_faces.py as,

add_faces.py

import cv2
import pickle
import numpy as np
import os
video=cv2.VideoCapture(0)
facedetect=cv2.CascadeClassifier('data/haarcascade_frontalface_default.xml')

faces_data=[]

i=0
name=input("Enter Your Name: ")

while True:
    ret,frame=video.read()
    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces=facedetect.detectMultiScale(gray, 1.3 ,5)
    for (x,y,w,h) in faces:
        crop_img=frame[y:y+h, x:x+w, :]
        resized_img=cv2.resize(crop_img, (50,50))
        if len(faces_data)<=100 and i%10==0:
            faces_data.append(resized_img)
        i=i+1
        cv2.putText(frame, str(len(faces_data)), (50,50), cv2.FONT_HERSHEY_COMPLEX, 1, (50,50,255), 1)
        cv2.rectangle(frame, (x,y), (x+w, y+h), (50,50,255), 1)
    cv2.imshow("Frame",frame)
    k=cv2.waitKey(1)
    if k==ord('q') or len(faces_data)==100:
        break
video.release()
cv2.destroyAllWindows()

faces_data=np.asarray(faces_data)
faces_data=faces_data.reshape(100, -1)


if 'names.pkl' not in os.listdir('data/'):
    names=[name]*100
    with open('data/names.pkl', 'wb') as f:
        pickle.dump(names, f)
else:
    with open('data/names.pkl', 'rb') as f:
        names=pickle.load(f)
    names=names+[name]*100
    with open('data/names.pkl', 'wb') as f:
        pickle.dump(names, f)

if 'faces_data.pkl' not in os.listdir('data/'):
    with open('data/faces_data.pkl', 'wb') as f:
        pickle.dump(faces_data, f)
else:
    with open('data/faces_data.pkl', 'rb') as f:
        faces=pickle.load(f)
    faces=np.append(faces, faces_data, axis=0)
    with open('data/faces_data.pkl', 'wb') as f:
        pickle.dump(faces, f)

  • Open terminal and type "python add_faces.py"
  • First it asks to "Enter your name:". Then enter our name. 
  • This will open your web camera. So, the process is again started. And this time, our face is surrounded by a rectangular box and timer will gets started. 
  • It will take 100 frames of our photo. Try to rotate face up to 360 degrees to take in various angles. 
  • After 100 seconds, camera will automatically close.
  • Now, automatically "faces_data.pkl" and "names_pkl" files are created in "data" folder. 
To see add_faces.py Click here
Now, create a file named "test.py" to test the input data given in add_faces.py file.

test.py

from sklearn.neighbors import KNeighborsClassifier
import cv2
import pickle
import numpy as np
import os
import csv

video=cv2.VideoCapture(0)
facedetect=cv2.CascadeClassifier('data/haarcascade_frontalface_default.xml')

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)


knn=KNeighborsClassifier(n_neighbors=5)
knn.fit(FACES, LABELS)

while True:
    ret,frame=video.read()
    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces=facedetect.detectMultiScale(gray, 1.3 ,5)
    for (x,y,w,h) in faces:
        crop_img=frame[y:y+h, x:x+w, :]
        resized_img=cv2.resize(crop_img, (50,50)).flatten().reshape(1,-1)
        output=knn.predict(resized_img)
        cv2.putText(frame, str(output[0]), (x,y-15), cv2.FONT_HERSHEY_COMPLEX, 1, (255,255,255), 1)
        cv2.rectangle(frame, (x,y), (x+w, y+h), (50,50,255), 1)
    cv2.imshow("Frame",frame)
    k=cv2.waitKey(1)
    if k==ord('q'):
        break
video.release()
cv2.destroyAllWindows()
  • Open a new terminal and type "python test.py"
  • This will open your web camera. So, the process is getting started. 
  • It will identify your face and displays your name.
  • Click "Q" to exit camera.
Now, write code as follows:

test.py

from sklearn.neighbors import KNeighborsClassifier
import cv2
import pickle
import numpy as np
import os
import csv

video=cv2.VideoCapture(0)
facedetect=cv2.CascadeClassifier('data/haarcascade_frontalface_default.xml')

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)


knn=KNeighborsClassifier(n_neighbors=5)
knn.fit(FACES, LABELS)

while True:
    ret,frame=video.read()
    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces=facedetect.detectMultiScale(gray, 1.3 ,5)
    for (x,y,w,h) in faces:
        crop_img=frame[y:y+h, x:x+w, :]
        resized_img=cv2.resize(crop_img, (50,50)).flatten().reshape(1,-1)
        output=knn.predict(resized_img)
        cv2.rectangle(frame, (x,y), (x+w, y+h), (0,0,255), 1)
        cv2.rectangle(frame,(x,y),(x+w,y+h),(50,50,255),2)
        cv2.rectangle(frame,(x,y-40),(x+w,y),(50,50,255),-1)
        cv2.putText(frame, str(output[0]), (x,y-15), cv2.FONT_HERSHEY_COMPLEX, 1, (255,255,255), 1)
        cv2.rectangle(frame, (x,y), (x+w, y+h), (50,50,255), 1)
    cv2.imshow("Frame",frame)
    k=cv2.waitKey(1)
    if k==ord('q'):
        break
video.release()
cv2.destroyAllWindows()
  • Open terminal and type "python test.py"
  • This will open your web camera. So, the process is getting started. 
  • It will identify your face and displays your name.
  • This time name and face are displayed in a rectangular box. 
  • Click "Q" to exit camera.
Create a folder named "Attendance". In that folder, create a file named "Attendance_20-08-2024.csv" file. 
Now, write code as follows:

test.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

def speak(str1):
    speak=Dispatch(("SAPI.SpVoice"))
    speak.Speak(str1)

video=cv2.VideoCapture(0)
facedetect=cv2.CascadeClassifier('data/haarcascade_frontalface_default.xml')

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)

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

knn=KNeighborsClassifier(n_neighbors=5)
knn.fit(FACES, LABELS)

imgBackground=cv2.imread("background.png")

COL_NAMES = ['NAME', 'TIME']

while True:
    ret,frame=video.read()
    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces=facedetect.detectMultiScale(gray, 1.3 ,5)
    for (x,y,w,h) in faces:
        crop_img=frame[y:y+h, x:x+w, :]
        resized_img=cv2.resize(crop_img, (50,50)).flatten().reshape(1,-1)
        output=knn.predict(resized_img)
        ts=time.time()
        date=datetime.fromtimestamp(ts).strftime("%d-%m-%Y")
        timestamp=datetime.fromtimestamp(ts).strftime("%H:%M-%S")
        exist=os.path.isfile("Attendance/Attendance_" + date + ".csv")
        cv2.rectangle(frame, (x,y), (x+w, y+h), (0,0,255), 1)
        cv2.rectangle(frame,(x,y),(x+w,y+h),(50,50,255),2)
        cv2.rectangle(frame,(x,y-40),(x+w,y),(50,50,255),-1)
        cv2.putText(frame, str(output[0]), (x,y-15), cv2.FONT_HERSHEY_COMPLEX, 1, (255,255,255), 1)
        cv2.rectangle(frame, (x,y), (x+w, y+h), (50,50,255), 1)
        attendance=[str(output[0]), str(timestamp)]
    imgBackground[162:162 + 480, 55:55 + 640] = frame
    cv2.imshow("Frame",imgBackground)
    k=cv2.waitKey(1)
    if k==ord('o'):
        speak("Attendance Taken..")
        time.sleep(5)
        if exist:
            with open("Attendance/Attendance_" + date + ".csv", "+a") as csvfile:
                writer=csv.writer(csvfile)
                writer.writerow(attendance)
            csvfile.close()
        else:
            with open("Attendance/Attendance_" + date + ".csv", "+a") as csvfile:
                writer=csv.writer(csvfile)
                writer.writerow(COL_NAMES)
                writer.writerow(attendance)
            csvfile.close()
    if k==ord('q'):
        break
video.release()
cv2.destroyAllWindows()
  • Open terminal and type "python test.py"
  • This will open your web camera. So, the process is getting started. 
  • It will identify your face and displays your name.
  • Your face is displayed in a rectangular box. 
  • Click "o" to take attendance.
  • Click "Q" to exit camera.
  • Attendance data is stored in "Attendance_20-08-2024.csv" file.
To see, test.py Click here

Create a file named "app.py"

app.py

import streamlit as st
import pandas as pd
import time
from datetime import datetime

ts=time.time()
date=datetime.fromtimestamp(ts).strftime("%d-%m-%Y")
timestamp=datetime.fromtimestamp(ts).strftime("%H:%M-%S")
df=pd.read_csv("Attendance/Attendance_" + date + ".csv")

st.dataframe(df.style.highlight_max(axis=0))

  • Open terminal and type "streamlit run app.py"
Fig: Streamlit app
  • This will open new web page.
Fig: Streamlit web page
  • Run test.py code
  • Then the process gets started. After clicking on "o", the attendance time will be noted. 
  • This can be shown in Streamlit web page.
  • Click "Q" to exit camera.
  • Also, Attendance data is stored in "Attendance_20-08-2024.csv" file.

app.py

import streamlit as st
import pandas as pd
import time
from datetime import datetime

ts=time.time()
date=datetime.fromtimestamp(ts).strftime("%d-%m-%Y")
timestamp=datetime.fromtimestamp(ts).strftime("%H:%M-%S")

from streamlit_autorefresh import st_autorefresh

count = st_autorefresh(interval=2000, limit=100, key="fizzbuzzcounter")

if count == 0:
    st.write("Count is zero")
elif count % 3 == 0 and count % 5 == 0:
    st.write("FizzBuzz")
elif count % 3 == 0:
    st.write("Fizz")
elif count % 5 == 0:
    st.write("Buzz")
else:
    st.write(f"Count: {count}")


df=pd.read_csv("Attendance/Attendance_" + date + ".csv")

st.dataframe(df.style.highlight_max(axis=0))
  • Open terminal and type "streamlit run app.py"
Fig: Streamlit app
  • This will open new web page.
Fig: Streamlit web page
  • Run test.py code
  • Then the process gets started. After clicking on "o", the attendance time will be noted. 
  • This can be shown in Streamlit web page.
  • Click "Q" to exit camera.
  • Also, Attendance data is stored in "Attendance_20-08-2024.csv" file.

Click Here for full code: Click here

Comments

Post a Comment

Popular posts from this blog

Comparative Analysis of Advanced Clustering Algorithms for Market Segmentation

  Comparative Analysis of Advanced Clustering Algorithms for Market Segmentation - A Case Study on Mall Customer Data Abstract This study conducts a comparative analysis of advanced clustering algorithms for market segmentation using Mall Customer Data. The algorithms evaluated include K-Means, Hierarchical Clustering, DBSCAN, Gaussian Mixture Models (GMM), Agglomerative Clustering, BIRCH, Spectral Clustering, OPTICS, and Affinity Propagation. Evaluation metrics such as Silhouette Score, Davies-Bouldin Score, and Calinski-Harabasz Score are employed to assess the clustering performance and determine the most suitable algorithm for segmenting mall customers based on their spending habits. Methodology The methodology involves several key steps: 1.      Data Collection: Mall Customer Data is obtained, comprising various demographic and spending attributes. 2.      Data Preprocessing: Data is cleaned, normalized, and prepared for cl...
Post a Comment
Read more

Enhanced House Price Prediction Using XGBoost: A Comprehensive Analysis with the Boston Dataset

 House Price Prediction Fig: Supervised Learning Types of Supervised Learning Fig: Types of Supervised Learning Boston House Price Prediction The Dataset used in this project comes from the UCI machine learning repository the data was collected in 1978 and each of the 506 entries represents aggregate information about 14 features of homes from various suburbs located in Boston. Fig: Boston Dataset Workflow Fig: Workflow Enhanced House Price Prediction Using XGBoost: A Comprehensive Analysis with the Boston Dataset Abstract The accurate prediction of house prices is a critical task in the real estate industry, aiding buyers, sellers, and investors in making informed decisions. This study explores the application of the XGBoost algorithm for predicting house prices using the Boston housing dataset. The model was evaluated using R-squared error and Mean Absolute Error (MAE) as performance metrics. The results demonstrate the model's effectiveness, with an R-squared error of 0.9116...
Post a Comment
Read more