www.ijcrt.

org © 2021 IJCRT | Volume 9, Issue 5 May 2021 | ISSN: 2320-2882

AI BASED ATTENDANCE MONITORING

SYSTEM
Kajal Chorghe1, Sharda Pawar2, Gayatry Sethy3, Vaibhav E Pawar4

Bharati Vidyapeeth College of Engineering (Navi Mumbai, Maharashtra, India)

ABSTRACT: In this project, face of a person is motion photos of the student when present in class
used for the aim of attendance making to analyse the student data how much time the
automatically. As Attendance plays an important student presents in class.
role for each college, universities and schools. A
Keywords: Face Detection, Face Recognition,
number of an automated biometric-based
OpenCV, Image Enhancement, Attendance
attendance system are projected within the
System
literature reviews, poor system models have
negatively affected their performance. Hence, a 1.INTRODUCTION
necessity to handle the various desires needed
with the development of an AI-based attendance As Attendance is an important factor for both the
observation system. Attendance observation student as well as the teacher of an academic
System is important in all organizations for organization. The student's attendance system
checking the performance of students and it is not using artificial intelligence idea mainly works
easy task to check each and every student is using the concept of Facial Detection, face
present or not. As we can see in all organization recognition system and Student’s attendance
attendance are taken manually by calling their roll observing system and applications. Maintaining
numbers or names and marked present in registers the attendance is very important in every and each
issued by the department heads as a proof and in groundwork for checking the performance of
some organizations the students wish to sign on students. Every organization has its own
these sheets which are stored for future references. technique. Face is considered as a primary key
In this system, we use are using the technique of feature to spot and speak with alternative peoples
utilization of face detection and recognition within the world because face considered as a
framework to endlessly recognize students going unique identity for every and each person. The
to class or not and marking their attendance by facial features will be unique to the other indusial.
comparing their faces with stored information in It processes overall facial structure, distances
database to match and marking attendance. This between color, ears, eyes, nose and mouth, etc.
facial biometric framework takes an image of a Except for computers, it's tough to analyse the
person using camera and distinction image with data therefore we tend to use the concept of
the stored image in the database at the time of Computer vision. The intention of using computer
enrolment and if it matches, marks the attendance vision technology to recognize the human features
and monitor the student performance endlessly. in a computer. Hence, this technique handles all
We may use the concept of artificial intelligence the problems that occurred in traditional system
to observe student attendance like capturing the and other bio metrics strategies.

IJCRT2105287 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org c668

www.ijcrt.org © 2021 IJCRT | Volume 9, Issue 5 May 2021 | ISSN: 2320-2882

2.LITERATURE REVIEW information/database server. This paper

introduces the efficient and correct methodology
(Marko Arsenovic et al) The developing of attendance within the classroom atmosphere
procedure is split into several necessary stages, that may replace the recent manual ways. This
together with the training dataset and methodology is secure enough, reliable and
augmentation, getting ready images and training accessible to be used. No need for specialized
DNNs and last but not least, integration into the hardware for putting in the system within the
prevailing system so as to check the proposed classroom.[5] There is a need to create such a
method. In this paper, a brand-new deep learning- system which will solve this problem and make a
based face recognition attendance system is user-friendly environment for the customer [10].
projected. The complete procedure of developing Also, human fingerprint considered as the most
a face recognition element by combining state-of trustworthy criteria for identification and it also
the-art ways and advances in deep learning is cannot be changed with time even up to the death
delineated. [1] of an individual [11].
(Louis Mothwa et al) It demonstrates how a
3.Proposed work
multicamera installation reduces the impact of
occlusion within the method of face detection, 3.1Architecture Diagram:
since higher face detection will increase the
accuracy of sophistication class attendance
observation. A wise model for student attendance
monitoring system supported on facial recognition
has been conferred in this paper. The real-time
face recognition with periodical updates of the
attendance info determines the presence of
students throughout the total lecture period. [2]

(RefikSamet et al) The system consists of three

layers: Application Layer, Communication Layer,
and Server Layer. A filtering system based on
Euclidean distances calculated by Eigenfaces,
Fisher faces, and LBP has been developed. The
fig.1 System Architecture
projected system eliminates the price for further
instrumentation, minimizes attendance-taking
time, and permits users to access the information 3.2Algorithms and Techniques Used
anytime and anyplace.[3]
1.Face Detection:
(Dr.B.R.Shunmugapriya et al)The group image of
Haar Cascade used for Face Detection in
the students is captured and shold on within the
OpenCV. Viola-Jones Face Detection techniques,
database. The individual faces of each student
popularly known as Haar Cascades. It is an object
area unit recognized by face detection algorithm
detection algorithm used to identify faces in
program. The project conferred the robust human image or a real live video. The algorithm is given
face recognition system supported on a lot of positive images consisting of faces and lot
discriminative robust native binary pattern and of negative images not consisting of any faces to
LDP for automatic attending system. [4] train on them. The model created from this
training is available at the OpenCV. Haar cascade
(Naveed Khan Balcoh et al) The system consists
detection is one of the most powerful face
of a camera that captures the pictures of the
detection algorithm not only used to detect faces,
classroom and sends it to the image enhancement
but also for eye,lips,licence number plates etc. we
module. After enhancement the image comes can access with OpenCV method.
within the Face Detection and Recognition
modules then the attendance is marked on the
IJCRT2105287 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org c669
www.ijcrt.org © 2021 IJCRT | Volume 9, Issue 5 May 2021 | ISSN: 2320-2882

4.RESULTS

Fig.4.1: Face capture

Fig.2 A sample of Haar features used in viola
and jones

2.Face Recognition:
 Apply PCA classifier
 Extract the features from detected faces
 Compare Features
 Put Student’s Attendance based on
recognized faces
 Steps Used in PCA Algorithm:

Step1: Acquire preparing set of 'N' number of

pictures at the first stage. In this technique
the pictures are of 92*112 pixels each.
Step2: Calculation of the Eigen face from the
"N" preparing set pictures keeping just couple of
M pictures that is relate to that of the highest
Eigen values. The M pictures indicate the "face
space”. At the point when new faces experienced,
the Fig 4.2: Registration Page

“Eigen faces" can be recalculated in like manner.

Step3: The relating distribution of the "M"
dimensional weight space for each known
individual is calculated by representing their
particular face pictures onto “face space".
Step 4: Compute set of the weights anticipating
the information picture or info picture to M
"Eigen faces".
Step5: Determine if the given picture is face
picture or not by checking to the closeness of Fig 4.3: Login Page
given picture or picture to "face space".

IJCRT2105287 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org c670

www.ijcrt.org © 2021 IJCRT | Volume 9, Issue 5 May 2021 | ISSN: 2320-2882

facedetection”E.Varadharajan,R.Dharani,
S.Jeevitha, B.Kavinmathi, S.Hemalatha.
5. “Face Recognition Based Attendance
System”Nandhini R,
Duraimurugan N,
S.P.Chokkalingam.
6. “Face Recognition Techniques - An
evaluation Study” Dr.Asmahan, M
Altaher.
Fig 4.4: Attendance report 7. “Face Recognition-Based Mobile
Automatic Classroom Attendance
Management System”RefikSamet,
Muhammed Tanriverdi.
8. “Algorithm for Efficient Attendance
5.CONCLUSION Management: Face Recognition based
We come to comprehend that there is approach”Naveed Khan Balcoh, M.
extensive form of methods, for example, Haroon Yousaf, Waqar Ahmad and M.
biometric, RFID based mostly and so forth which IramBaig.
are time intense and non-efficient. So, to 9. “Automated Attendance Management
overcome, this on the top of framework is that the System Based On Face
higher and reliable resolution from each RecognitionAlgorithms”Shireesha
perceptive of our time and security. In this way Chintalapati, M.V. Raghunadh.
we've accomplished to feature to a reliable and 10. Bhawana S. Dakhare, Amish Khatu,
effective participation framework to differentiate Hrushikesh Baliar Singh, Aniket Gupta,
faces in classroom and recognize the faces "Visual E-commerce Application using
accurately to mark the attendance. The scope of Deep Learning", International Research
the project is that the system on which the Journal of Engineering and Technology
software is installed, i.e., the project is developed (IRJET) Volume: 07 Issue: 03 | Mar 2020
as a desktop application, and it'll work for a 11. Patil, V., Ingle, D.R. An association
particular institute. However afterward, the between fingerprint patterns with blood
project can be changed to control it online. group and lifestyle based diseases: a
review. Artif Intell Rev 54, 1803–1839
(2021). https://doi.org/10.1007/s10462-
020-09891-w
6.REFERENCES

1. “Face Detection and Recognition for

Automatic Attendance System Using
Artificial Intelligence Concept” Mr
PradeepaM , H P Mohan Kumar.
2. “FaceTime – Deep Learning Based Face
Recognition Attendance System” Marko
Arsenovic, SrdjanSladojevic, Andras
Anderla, Darko Stefanovic.
3. “Conceptual Model of the Smart
Attendance Monitoring System Using
Computer Vision “Louis Mothwa, Jules-
Raymond Tapamo, Temitope Mapayi.
4. “Automaticattendance management
system using
IJCRT2105287 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org c671