Volume 8, Issue 6, June – 2023 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Face Recognition Attendance System

Kesthara. V1 Rashmi R2; Bhaskarareddy N C2;
1
Assistant Professor, Electronics and Communication Monika Kumari2; Meera M2
2
Department, Dr. Ambedkar Institute of Technology, Student, Electronics and Communication Department, Dr.
Bengaluru, Karnataka, India Ambedkar Institute of Technology,
Bengaluru, Karnataka, India

Abstract:- The Attendance system is an important session is going on, faces will be detected from the live
administration department of any organisation, like streaming video of the camera in the classroom. The faces
schools, colleges, hospitals, offices, companies, or detected will be compared with each student's images stored
working areas. The production, performance, wages, in the database. If the match is successful, attendance will
and salaries of the employees or workers and the overall be recorded for the respective student. At the end of each
statistics of an organisation are mainly dependent on class, a list of absentees will be mailed to the faculty
parameters such as hours, days, and the number of handling that particular session of class.
people working. Monitoring attendance is highly
essential, and it interconnects huge amounts of data as it Reference [13] A camera is installed at the stage,
concerns the data analysis and the information of any middle, or centre of the class room at a height so as to get a
company, institution, or organisation, as records of it are full view of the class until the last bench, covering all the
important. This paper mainly focuses on using the students in the camera. When students have settled down,
technology of face recognition to improve the attendance the camera takes a picture and starts the face detection
system of students in a college or institution. The process using the Viola Jones algorithm. For face
attendance of students is the major factor that influences recognition functions, which are mentioned in this project, a
the student’s exam eligibility criteria and performance neural network architecture with LBPH (Local Binary
evaluation. Pattern Histogram Algorithm) is used. After all the
algorithms are executed, face recognition is done, and the
Keywords:- Facial Recognition, Open CV, Machine programme designed will automatically create a folder in
Learning, Technology, Python. the database for having recognised the students.

I. INTRODUCTION The reference [14] system's proposed aim is to capture

each student's face and store the face images in the database
On a day with seven-hour classes and considering 60 for future workings. The system in this paper uses
students, the students attendance is noted by the teacher in Convolutional Neural Networks algorithm. A convolutional
every class hour, which equates to marking the attendance neural network (CNN, or ConvNet) is a type of deep neural
seven times a day. The approximate time taken by the network that is used to analyse visual imagery. The student's
teacher to call the students names and mark their attendance face is captured in such a way that all of the student's facial
is 10 minutes. This method of taking attendance takes away traits, location and posture, are detected.
10 minutes of each one-hour class, and the remaining time
left for the teacher or lecturer to skill the students is 50 The reference [15] proposed system improve the
minutes, which has to be used judiciously. This 10 minutes attendance management system using of our unique
of each hour-long class causes interference with the flow of characteristics of their face. For the purpose of confirmation
thoughts, and gradually academic productivity drops. and documentation face acknowledgment technique is used.
Therefore, using face recognition in the field of attendance In this system we can use Haar Cascade method. A Haar
management systems will be helpful as there will be no use Cascade is used in image recognition and image processing
of books or paper and time will be reduced, which can be that is specially designed for pixel data.
utilised for teaching more to the students. Face recognition
is a contactless method of marking attendance. By In our proposed system we also use Haarcascade
developing and implementing automation in various fields, algorithm for face recognition with the feature of GUI for
we will encourage people to use the latest technologies and the teacher to enter the respective class details and upload
have a better future. the class photo to detect the face and mark attendance. We
have taken the database of the students and trained them
II. LITERATURE SURVEY many times so that we can get more accuracy and detect
faces correctly. So this system eliminates the existing
A proposed method, as outlined in reference [10], traditional way of taking attendance by calling out the
explains that the students of each and every class will have student names and then marking the attendance.
to first register themselves by entering their details with
their different-angle images of faces captured, which are
stored in a database for face recognition. When the class

IJISRT23JUN2507 www.ijisrt.com 2265

Volume 8, Issue 6, June – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
 Objectives face recognition technology is the creation of a database for
the undergraduates in a class. In this paper, we have
 To reduce the consumption of time take during designed a Python programme to create the student's
attendance marking in the class database. When the Python programme is run, the camera
 To automize the attendance functioning without much will capture the student's face from different angles and with
human interference. different expressions. A student's database will have 20 to
 To increase the productivity graph by managing the 25 different images of various angles of the face, and they
attendance using technologies. are stored in a folder by the student's name. In a class of 60
 To increase the speed of analysing the students students, the database is created with 60 folders with the
performance data and academic results by taking students names, or USN.
attendance using the modern technique of face
recognition. When we consider the actual classes conducted in
 To introduce the latest technology in order to manage colleges, there are separate departments and eight semesters
attendance. for four years. Each semester has six to seven subjects. The
 To eliminate proxy attendance when the attendance is attendance of the students is marked for each subject every
taken. day. Under each subject, all the students' databases for that
subject class are saved.
III. METHODOLOGY
Here folders are created for each department,
A. Creation of Database : semester, and subject, and accordingly, the student database
The very first procedure to be followed in order to folder is created in the hierarchical way shown in Fig.1. The
design an attendance management system with the help of letters A, B, and C stand for the students’ names.

Fig 1 Database Creation Structure

B. Training the Database : C. Testing of the Database

Training the newly created database is the next step in The trained data is then tested for efficiency and
the process. Here, training Python code has been designed accuracy. This paper introduces a GUI for the teachers to
and run for the purpose of training the database of the use on the desktop to enter details like year, semester, and
students for face recognition. Training the face dataset subject to take attendance of the students coming under that
means training the machine learning model to recognise, particular subject class. After entering the details, the
discover, and learn the patterns. Training the database is the students will be facing the camera, and the face detection is
major data learning process, as it involves algorithms to done through the testing code written in Python. An
discover patterns, make observations, and take decisions. attendance Excel sheet is then created for the particular
Here, the function of determining and learning the optimal subject class with the markings of students present and
combinations of variables is done during the development of absent. In this GUI part, we have included another option
the model by training the data. The data features are for uploading a photo. This photo is of the students sitting in
identified, and the relationships are found in order to predict the class and was taken through a mobile. All students will
the desired outcome. The model that has been designed is have to focus the phone camera while the photo is taken.
for analysing the dataset repeatedly over and over again for Through this photo, faces can also be detected, and
deeply understanding its characteristics and for adjusting by attendance can be marked for the class by uploading the
itself to get a better performance. image using other options designed in the GUI part. The

IJISRT23JUN2507 www.ijisrt.com 2266

Volume 8, Issue 6, June – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
diagram below shows the flow of functions of the Android app to be used by teachers for entering the details,
attendance management system using the technology of face and the particular camera gets activated, detects the faces,
recognition and identification proposed in this paper. In and marks attendance automatically.
real-time applications, the GUI part can be extended to an

Fig 2 Block Diagram

 Open CV  To Display an Image:

OpenCV is a library developed by Intel that is used by Use the function cv2. imshow(),cv2.destroy Window
many people working in different companies and students of () and cv2.waitKey(0)
different institutions for learning and working purposes, as
it is open-source computer vision and machine learning  Python IDE
software. It facilitates being used as a common Python IDEs are software platforms that are primarily
infrastructure for all kinds of applications that are related to useful for programmers and developers working on
working with computer vision and its associated fields. In application development, as they provide a comprehensive
our proposed system of attendance management, it is used set of tools, settings, or options for software development
to speed up the process of real-time machine recognition of with the help of a single product, specifically in the Python
images, objects, and video processing applications. In our programming language. Python IDEs are built to make it
paper, the Haarcascade features for the face detection easy and flexible to work with specific application purposes.
technique are worked upon. Python can work with different platforms, such as
Windows, Mac, Linux, Raspberry Pi, etc. Python uses a
 To Initialize the Camera : very simple syntax that is almost similar to the English
Use the function cv2.Video Capture(). The picture must language and can be understood by anyone. Python has a
have a complete path specified syntax that minimises the use of more code lines by
developers in order to write programmes when compared to
 To Capture Image From Camera : other programming languages. Python is a programming
Use the function cv2.imread("Image.jpg") language that can be run on an interpreter system, which
means that the code can be executed as soon as it is written.
 To Convert the Captured Colour Picture to Grayscale: The prototyping of the designed application can be obtained
Use cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) very quickly.

 To Detect Students Faces in the Grayscale Image:

Use face_cascade. Detect Multi Scale (img, scale
Facto r=1.1, minNeighbors=5)

IJISRT23JUN2507 www.ijisrt.com 2267

Volume 8, Issue 6, June – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
IV. RESULTS Then for the real time purpose the laptop camera gets
on and it will detect the face of the person visible to the
The below Fig.3 shows the photo of the GUI for camera which is shown in Fig.5.
selecting option for creation of database, training of image,
uploading image and real time face recognition testing.

Fig 6 GUI to Detect Face through Image

Fig 3 GUI to Create, Train, Detection using After the selecting the option of uploading image from
Real Time or Image Fig.3 the screen appears as shown in Fig.6. Here the teacher
fills the required details of year, department, semester and
subject. Then the submit button is pressed to load the
entered details and after that test image button is clicked.

Fig 4 GUI for Real Time Face Detection

After the selecting the option of real time face Fig 7 Class Photo for Detection and Marking Attendance
recognition the screen appears as shown in Fig.4. The
teacher fills the required details of year, department, The photo of the class students is taken and uploaded
semester and subject. Then the submit button is pressed to for detection of faces when test image button is selected.
load the entered details and after that real time button is The Fig.7 shows the class photo uploaded for the detection.
clicked.

Fig 5 Capturing the Students Face through Camera Fig 8 Database Created of the Student

IJISRT23JUN2507 www.ijisrt.com 2268

Volume 8, Issue 6, June – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
The Fig.8 shows the folder created with the name of REFERENCES
student in which the pictures are stored as database that will
be used to compare the faces that are cropped from the [1] Bhumika G. Bhatt, Zankhana H. Shah “Face Feature
uploaded image. Collecting of database is done at the Extraction Techniques: A Survey”, National
starting of the course and remain same until the completion Conference on Recent Trends in Engineering &
of the same. Technology, 13-14 May 2017.
[2] G. Yang and T. S. Huang, “Human face detection in
complex background,” Pattern Recognition Letter,
vol. 27, no.1, pp. 53-63, 2018.
[3] Kar, Nirmalya, et al. "Study of implementing
automated attendance system using face recognition
technique." International Journal of computer and
communication engineering 1.2 (2019): 100.
[4] Roshan Tharanga, J. G., et al. "Smart attendance
using real time face recognition (smart-fr)."
Department of Electronic and Computer Engineering,
Sri Lanka Institute of Information Technology
(SLIIT), Malabe, Sri Lanka (2020)
[5] Selvi, K. Senthamil, P. Chitrakala, and A. Antony
Jenitha."Face recognition based attendance marking
system." Corresponding Author: S. Rajkumar (2021).
[6] Joseph, Jomon, and K. P. Zacharia. "Automatic
Fig 9 Attendance Stored in Excel Sheet attendance management system using face
recognition." International Journal of Science and
After face detection using real time or through test Research (IJSR) 2.11 (2022): 327-330.
image the attendance of the students is marked and stored in [7] M. Turk, A. Pentland, Eigenfaces for Recognition,
excel sheets as shown in above Fig.9. Journal of Cognitive Neurosicence, Vol.3, No. 1,
Win. 2017, pp. 71-86
In our experiment we have taken database of around [8] Discriminant analysis for recognition of human face
27 students and after testing the uploaded image it has been images by Kamran Etemad and Rama Chellappa.
able to detect faces of around 20 to 25 students from that [9] Sanjay singh et.al, “A robust skin color based face
test image. detection algorithm”, Tamkang Journal of Science
and Engineering vol.6, no.4,pp227-234, 2018.
V. CONCLUSION [10] "Face Recognition based Attendance Management
System” by Smitha, Pavithra S Hegde in May 2020
In this proposed model or system, the attendance [11] Facial expression recognition using a combination of
marking work and management have been made easier by multiple facial features and support vector machine"
using technology for their development, like face by Hung-Hsu Tsai and Yi-Cheng Chang (July, 2018)
recognition. Attendance is a major factor that influences the [12] “An Evaluation of Face Recognition Algorithms and
student's performance, results, and exam eligibility. Accuracy based on Video in Unconstrained Factors"
Therefore, with the traditional way of taking attendance and by Phichaya Jaturawat and Manop Phankokkruad
marking it on a book or paper, more time is consumed to (2016)
analyse each and every student’s attendance and to use that [13] “Smart Attendance system using Face recognition”
parameter to calculate the student's result or performance by Ghalib Al-Muhaidhri , Javeed Hussain (2019)
analysis, and also more work is involved to get attendance [14] “Automated Attendance System Using Face
recorded onto the computer. Recognition” by Korukanti Ashritha (2022)
[15] “ Attendance Management System using face
This paper shows the attendance system being recognition” by Rucha, Soundarya, Gaurav, Ashwini
automated and getting recorded in the computer with a (2021)
reduction in time, and further, the attendance of students can
easily be linked to anywhere, like the result sheet, academic
review, and for analysing performance graphs.

IJISRT23JUN2507 www.ijisrt.com 2269