See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/371788782

A REVIEW PAPER ON ATTENDANCE MANAGEMENT SYSTEM USING FACE

RECOGNITION

Conference Paper · June 2023

CITATIONS READS

0 457

3 authors, including:

Ayush Srivastava
Galgotias University
2 PUBLICATIONS 2 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ayush Srivastava on 23 June 2023.

The user has requested enhancement of the downloaded file.

 A REVIEW PAPER ON ATTENDENCE MANAGEMENT
SYSTEM USING FACE RECOGNITION
Ayush Srivastava¹ Pragya katiyar² Mohini Chauhan³

Guide name: Vinod Kumar, Assistant Prof.

School of Computer Science & Engineering
Galgotias University, Greater Noida

ABSTRACT
The present time attendance marking system has developed into a system that is challenging interesting, and
reliable. In a large classroom, when many students attend the class, it is difficult to record each student's
attendance. In the current research, various attendance management systems have been implemented for use.
However, there are still problems with the recognition of faces in the attendance management system, which
allow more research to be done to resolve problems and make the system more effective. This paper will conduct
a review of previous research that various researchers have completed for their research papers. In addition to
reviewing the literature on previous research or related works, this study provides an in-depth review of the
major components, discussions, & suggestion for future works.

Keywords: - Histogram Oriented, Face Recognition, and PCA Automatic system, AMS, and gradient Haar Cascade.

others. These technologies, however, operate in a

1. INTRODUCTION queue, which could take longer and be more
obtrusive. In this research, we suggested a system
In recent years, a face recognition technology has that detects students' faces in real-time classroom
developed, FACE Recognition has gained a lot of camera and records their attendance if the detected
attention and increased is importance. Also, it is an face is recognized in the database. Compared to the
important use for analysis of pictures, and creating a old ways, the new method will take less time. The
machine-learning face recognition system that process of face recognition requires multiple steps,
accurately recognizes human faces is extremely including capture, extraction, comparison, and
challenging. The issue of manual attendance is matching. Each process has the following functions:
difficult to solve and takes a lot of time. This system Capture is the first step in the enrollment process for
is applied by various government agencies as well as the system, and it involves taking a picture. The next
educational institutions to record attendance. By step, extraction, is used to identify or remove the
analyzing real-time captures, face recognition is a specific object from the face. Then, during the
biometric approach to recognizing a person. Face comparison step, fresh data are used to compare with
recognition technology is growing into a universal enrollment or database samples. The final step is
biometric approach since, compared to other matching. Based on the extraction and comparison
biometric possibilities, it takes almost no user effort. steps, the system will be able to determine whether
Additionally, the faculty must individually are or not the new face matches the original face.
calling each student's name in order to record
attendance, which could take up to five minutes for
the whole class. This takes a lot of time. There is a
possibility that a proxy will show up. As a result of
this, several institutions started using a variety of
additional methods for maintaining attendance,
Including Radio Frequency Identification (RFID)
[3], iris recognition [4], fingerprint recognition, and

1
 3. PROPOSED SYSTEM :
Before photos of the students are taken and added to
the records, each student in the class must register and
submit all the required data. Each lesson's live-
streaming video from the classroom will show the
same faces. The image data from the record will be
compared to the recognized faces. If a match is found,
the relevant student's attendance will be recorded. A
list of absentees for each session will be mailed to the
faculty member in charge of it.
The suggested system's system architecture is shown
Fig : The operates process of face recognition system below:

2. LITRETURE SURVEY

A design for an attendance system based on

computers was offered by the authors of [1]. Face
recognition and Radio Frequency Identification
(RFID) are used in combination as part of the plan to
identify and count authorized students as they enter
and exit the classroom. A proper record is kept for
each student who has registered with the system.
Additionally, the system maintains records of each
student's registration information for a particular class
in the attendance record and sends the required data
when needed. An iris biometric attendance system has
been created and put into use by the investigators of
the research [2]. Participants had to fill out an
individual iris form and register with their personal
information at first. By snatching a picture of each
student's eye, recognizing their iris, and searching for
a match in the provided data, the device immediately
counted attendance for the class.
An intelligent attendance marking and management
system based on face recognition is presented in Fig: System Architecture
another work [3] by Jenif D. Souza. This device
automatically recorded each student's attendance by Usually, there are four parts to this procedure.
taking a photo of them while they were in class using
the camera. The histogram algorithm is used by this 1. Datasets Creation:
system. The histogram approach is employed with the Using a webcam, images of students are recorded. We will
intention of identifying faces. The face image is take several pictures of a single student from various
converted into a series of photos using this technique. angles and movements. Pre-processing is done on these
Histograms are used to accurately recognize faces. photos. To obtain the Region of Interest (ROI), the photos
With this system, the time-consuming issue is are resized. This ROI will then be used in the recognition
resolved.. procedure. The photos must then be resized to a specific
In other paper [4] 2019 Nandhini R. introduced Attendance pixel position after being cropped. Following that, these
System based on face recognition. The system records photos will be transformed from RGB to grayscale. The
student images, analyses it into smaller frames, and stores names of the appropriate students will be added to these
them in a file system. Face detection is also accomplished images before they are saved in a folder.
using the Convolution Neural Network (CNN) technique.
Accuracy and speed are improved by this system.

2
2. Face Detection: recognition process, the spreadsheet will display the
Here, face detection is done with OpenCV and the Haar- recognized faces as present and the remaining faces as
Cascade Algorithm. To use the Haar Cascade method for absent, with the sending of the absentee list to the
recognizing individuals, it must first be trained to respective faculties. Every month's summary, teachers
recognize human faces. The term for this is feature will receive an updated attendance report.
extraction. The xml format
filehaarcascade_frontalface_default provides the training COMPARISON OF FACERECOGNITION
data for the haar cascade. TECHNIQUES:
The Haar features in shown fig 1.2.
The technical record on the face recognition-based
attendance management system has been modified.
We looked at different papers, and they utilized fresh
approaches or methods. The chart of instances for
several publications is shown below.
To develop our research paper, we read more than
research papers. This paper helped us to understand
different concepts and challenges, as well as how to
make our research more efficient and less time-
consuming.

Fig: Haar figure

3. Face Recognition:
The method of detecting faces can be divided into
three parts.create training data, train a face recognizer,
and make predictions. The images included in the
dataset will serve as the training data in this case.
They will be given an integer label indicating which
student they relate to. Face recognition software is
then used to these pictures. This system uses the
Local Binary Pattern Histogram as a face recognizer.
The list of local binary patterns (LBP) for the entire
face is first created. Following the the decimal system
of these LBPs, histograms of all the decimal values
are created. One histogram will be created at the
conclusion for each image in the training data.

4. Attendance updation:
After the face recognition process, the spreadsheet
will display the recognised faces as present and the
remaining faces as absent, with the list of absentees
being mailed to the respective faculties. After the face

3
Primary Creation and Training: Attendance Marking Phase
We will be training the pictures in their order using
the LBPH algorithm while always keeping an eye on The attendance is registered with the time and date for
their histogram number and comparing stored and that lecture and saved if the uploaded image matches
trained against photographs of identifying those the image stored in the database during the process
present. The first site containing student photos was that comes after the identification phase. After that, an
created by taking real-time live video Readers, excel spreadsheet is generated.
separating the video into thirty structures,
transforming to grey scale, and keeping only the face
students as pictures. Software for categorising video
frames for Open-CV.

Fig: Attendance being marked excels sheets

Block Diagram of face Recognition:

Fig: Image stored in database

4
Result: Conclusion:

The face recognition attendance management system Before the project's release. The traditional method of
is amazing user-friendly and performs effectively taking attendance has many mistakes, which has
under short schedules. Since this is an automated caused major problems for most schools. The
system, if an administrator generated a student profile proposed Attendance Management System Using
in the database just once, it would instantly use that Face Recognition is a more efficient method of
profile as many times as necessary during the face monitoring student attendance in classrooms and
detection and identification process. In this system, other situations. Despite the wide range of
Haar Cascade techniques are used. technologies currently easily accessible, including
To begin using this system, the administrator must biometrics and other methods, facial recognition is the
first build a profile for each student that includes their most accurate.
name, roll number, department, and other academic The system requires to develop a powerful face
information. recognition-based class attendance system. The
proposed system will be capable of recording
attendance using face identification. It will use a
webcam to detect faces and then identify them.
Following recognition, it will update the attendance
record and note the recognized student's attendance.

Reference:
[1] Jenif D Souza, Jothi S, Chandrasekar A,
“Automated Attendance Marking and Management
System by Facial recognition using Histogram” in
ICACCS 2019.
[Ref: https://ieeexplore.ieee.org/document/8281895]

[2] Nandhini R, Duraimurugan N, S.P Chollalingam “

Face Recogn:\
[Ref:
https://www.ijeat.org/wpcontent/uploads/papers/v8i3
S/C]

[3] E Varadharajan , R Dharani , S.Jeevitha, B

Kavinmathi, S. Hemalatha “ Automatic Attendance
Management system using face detection” at ICGET
2016. 2020 Department of Information Technology.
[Ref:
When everything is completed, create the student's https://ieeexplore.ieee.org/abstract/document/7916753
marked attendance-shit in Excel crap format. ]
If some entered data is not in our dataset then it is
placed in the dataset as an unknown data. [4] Poornima S,Sripriya N , Vijayalakshmi B,
The attendances management system is able to Vishnupriya P “ Attendance monitoring system using
recognize every image with various angles and facial recognition with audio output gender
brightness. classification “ in ICCCSP 2017.
Real-time attendance marking is done by identifying [Ref: https://ieeexplore.ieee.org/document/7944103]
student face.

[5] Salim, Omar Abdul Rhman, Rashidah Funke

Olanrewaju, and Wasiu Adebayo Balogun. "Class
attendance management system using face
recognition." 2018 7th International Conference on

5
Computer and Communication Engineering (ICCCE). [13]. Shrey Bhagat, Vithal Kashkari , Shubhangi
IEEE, 2018. Srivastava , Ashutosh Sharma, “Face recognition
attendance system” 2022, Available:
[6] Radhika C.Damale, [Ref: https://www.ijraset.com/research-paper/face-
Prof.Bageshree.V.Pathak. “Face Recognition Based recognition-attendance-system]
Attendance System Using Machine Learning
Algorithms." Proceedings of the Second International [14]. Smitha, Pavithra S Hegde, Afshin, Face
Conference on Intelligent Computing and Control recognition based attendance management system”
Systems (ICICCS 2018) IEEE Xplore Compliant 2020, Available:
Part Number: CFP18K74- [Ref:
https://www.researchgate.net/publication/341876647]
[7] Omar Abdul Rhman Salim, Rashidha Olanrewaju
,Wasiu Balogun “Class attendance management [15]. Harihara Santosh Dadi, Gopala Krishna Mohan
system using face recognition” in ICCCE 2018. Pillutla, “Improved face recognition rate using HOG
[Ref: features and svm classifier” 2016
http://www.researchgate.net/publication/329067820_ [Ref:
Class] https://www.researchgate.net/publication/305709603]

[8] Shreyak Sawhney, Karan Kacker, Samayak Jain,

Shailendra Narayan ,Rakesh Garg “Real Time Smart
Attendance system using face recognition techniques
“ in international conference on cloud computing data
science and engineering 2019.
[Ref:
https://ieeexplore.ieee.org/abstract/document/8776934
]

[9]. Bharath Tej Chinimilli, Amrita Vishwa

Vidyapeetham, Vihas Reddy Kaipu, Amrita Vishwa
Vidyapeetham, Anjali T, Akhil Kotturi, Jathin Varma
Mandapati, "Face recognition based attendance
system using Haar cascade and local binary pattern
histogram algorithm" 2020,
[Ref: https://ieeexplore.ieee.org/document/9143046]

[10]. Smit Hapani, Nikhil Parakhiya, Nandana

Prabhu, Mayur Paghdal, "Automated attendance
system using image processing" 2018,
[Ref: https://ieeexplore.ieee.org/document/8697824]

[11]. Sikandar Khan, Adeel Akram, Knight Usman,

“Real time automatic attendance system for face
recognition using face API and OpenCV” 2020,
[Ref:
https://link.springer.com/article/10.1007/s11277-020-
07224-2]

[12]. Shubhobrata Bhattacharya, Gowtham Sandeep

Nainala, Prosenjit Das and Aurobinda Routray,
"Smart attendance monitoring system (sams): a face
recognition based attendance system for classroom
environment" 2018, Available
[Ref: https://ieeexplore.ieee.org/document/8433537]

6
7
 8

View publication stats