Real Time Web-Based Smart Attendance System Using AI

ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)
Volume 3, Issue 2, April 2023

Impact Factor: 7.301
Real Time Web-Based Smart Attendance System

using AI
Maaz Patel1, Sanskruti Patil2, Rishikesh Khade3, Abhishek Sananse4 , Prof. Pragati Deole5
Students, Department of Computer Engineering1,2,3,4
Professor, Department of Computer Engineering5
Smt. Kashibai Navale College of Engineering, Pune, Maharashtra, India
Abstract: The performance of automatic face recognition (AFR) technology has improved in recent years.
As a result, these systems are frequently employed for security and business purposes. a face recognition
programme that runs automatically in the background while a university tracks employee attendance.
Hence, Smart Attendance with Real Time Face Recognition might be a global solution for managing
personnel on a daily basis. The work is incredibly challenging since the critical temporal background
subtraction during an image remains a challenge. Real-time face detection is done using easy, quick
Principal Component Analysis, which has a high rate of success in identifying the faces found. As a result,
we have developed a module in which an employee's attendance will be recorded automatically by
recognising their face using face recognition technology. Their face will be saved in the database after this
one-time enrollment process. Since enrolling a face is a one-time activity, we need a mechanism. Every
employee will have a unique employee ID that is their own roll number. Every employee's presence will be
updated within a database. Results of the proposed system demonstrated that it performs better than the
manual attendance system. The marking of attendance follows employee identification. This application
provides the customer with considerably more solutions and accurate leads than the conventional
attendance and leave.
Keywords: Attendance, Face Recognition, CNN (Convolutional Neural Network)
I. INTRODUCTION
Keeping up with the participation is critical through and through the organizations for really taking a look at the
presentation of representatives (4). Each establishment has its own strategy during this respect. Some are gauging
participation physically utilizing the old paper or record based approach and a couple of have embraced strategies for
programmed participation utilizing some biometric procedures. Be that as it may, in these strategies representatives
need to anticipate while in causing a line at time they to enter the workplace. Numerous biometric frameworks are
accessible yet the key validations are same is every one of the strategies. Each biometric framework comprises of
enrolment process during which extraordinary highlights of an individual is put away inside the data set then there are
cycles of ID and check. These two cycles think about the biometric element of a person with recently put away format
caught at the hour of enrolment. Biometric layouts are frequently of the many sorts like Fingerprints, Eye Iris, Face,
Hand Math, Mark, Walk and voice. Our framework utilizes the face acknowledgment approach for the mechanized
participation of workers inside the workplace room climate without representatives' intercession (2). Face
acknowledgment comprises of two stages, in drive faces are identified inside the picture then these recognized
appearances are contrasted and the data set for confirmation. Various strategies are proposed for face identification for
example Ada Lift calculation, the Float Lift calculation, the S-Ada Lift calculation Backing Vector Machines (SVM),
and accordingly the Bayes classifier. The proficiency of face acknowledgment calculation are frequently expanded with
the quick face recognition calculation. In every one of the above strategies SURF is best. Our framework used this
calculation for the discovery of appearances inside the workplace room picture. Face acknowledgment strategies are
frequently Partitioned into two sorts Appearance based which use surface highlights that is applied to entire face or a
few explicit Locales, other is Component based which utilizes mathematical elements like mouth, nose, eyes, eye
foreheads, cheeks and Connection between them. Measurable apparatuses like Straight Discriminant Examination
Copyright to IJARSCT DOI: 10.48175/IJARSCT-9112 124
www.ijarsct.co.in
IJARSCT

(LDA), Head Part Investigation (PCA), Portion Strategies, and Brain Organizations, Eigen-faces are utilized for
development of face layouts.
II. LITERATURE SURVEY

Proposed how of gauging participation utilizing educator's cell phone has been introduced during this paper which is
paperless, speedy, and precise. An application programming introduced inside the teacher's cell phone empowers it to
scrutinize understudies' cell phone by means of Bluetooth association and, through move of understudies' cell phones'
Media Access Control (Macintosh) locations to the educator's cell phone, presence of the researcher are frequently
affirmed. Additionally, point by point record of an understudy's participation can likewise be created for printing and
documenting, if necessary [1].
In this Paper they planned The Face Acknowledgment Terrific Test (FRGC) is intended to understand this exhibition
objective by introducing to scientists a six-explore challenge issue close by information corpus of 50,000 pictures. The
information comprises of 3D sweeps high goal still symbolism taken under controlled and uncontrolled circumstances.
They conquered the test issue, information corpus, and presents pattern execution and fundamental outcomes on normal
measurements of facial symbolism [2].
Proposed a remote iris acknowledgment participation the executives framework which was planned and carried out
utilizing Daugman's calculation (Daugman, 2003). This framework based biometrics and remote method settles the
issue of false participation and subsequently the trouble of laying the relating organization. They expected that this
framework make the clients' attendances all the more effectively and successfully [3].
CCTV film is generally used in the court to help picture the wrongdoing being referred to and to help recognize the
guilty party. Tragically, the greater part of reconnaissance cameras produce such low quality pictures that the
assignment of recognizing people are frequently incredibly troublesome. This study pointed toward deciding if the
undertaking of recognizing the guilty party in CCTV film was one which a jury ought to be able to attempt to, or
whether master proof would be useful in such cases. The capacity of potential jury individuals, the general public, was
tried by requesting that members assume the part of a jury part through a web study. Potential jury individuals saw
CCTV during which a reproduced offense occurred, and were hence approached to match actually pictures of a litigant
to the guilty party to embrace to decide whether they were able and guaranteed about making a judgment on whether
the respondent carried out the wrongdoing.
Factors, for instance, age, sexual direction and calling of the potential jury people were considered, similarly as the kind
of bad behavior did, to develop expecting that these accept any part in decision by potential jury people. These factors
didn't appear to expect an immense part; regardless, conviction was moreover explored and it ended up being
outstandingly evident that this was a component that ought to be pondered while choosing the essential for ace
responsibility in facial assessments [4].
The particle smoothing out with flexible consolidated course based histogram improvement technique (PACDHE) for
additional fostering the continuous video quality. At first the accounts are accumulated, each coming edge has been
penniless down and noise present in the video frame is discarded by applying the non-divisional center channel.
Starting there forward, nature of persistent video is overhauled iteratively by examining each pixel present in video
frames using the high level wellbeing and consolidated scattering work. This cycle is repeated continually until to
overhaul the continuous video frames separation and quality. Then, the introduction of the structure is explored by
using CV electronic video informational collection and the efficiency is dissected to the extent that apex sign to
upheaval extent (PSNR), Outright Mean Splendor Mistake (AMBE) and Entropy. The test results of PSO are
differentiated and genetic estimation based technique and saw that PSO beats the GA approach and the ongoing
histogram balance approach and the ongoing histogram evening out approaches [5].
In human machine affiliation customized talk feeling affirmation is yet troublesome yet huge task which gave close
thought in stream research locale. As the gig of talk is a climb in human PC interface. Talk is engaging and strong
medium because of its couple of features imparting disposition and sentiments box talk is feasible. Here audit is
overseen using Gaussian mix model and Secret Markov model classifiers used for ID of 5 fundamental energetic states
of speaker's as shock, delight, hopeless, shock and fair-minded. in this paper to see sentiments through talk various
features, for instance, prosodic components like pitch, energy and ghost components, for instance, Mel repeat cestrum
www.ijarsct.co.in
IJARSCT

coefficient were isolated and considering this components energetic portrayal and execution of collection using
Gaussian blend model and Secret Markov Model is discussed.[6]
III. METHODOLOGY
In appropriate settings, such as colleges and offices, the suggested method is utilised to manage attendance by using
face recognition to take attendance. The Raspberry Pi Camera Module V2 linked to a Raspberry Pi3 can also be used to
build the system architecture, which can be installed where people enter the office or classroom. To record video, utilise
the Camera Module. which human face photos are extracted. During face recognition, OpenCV library files are used to
automatically verify the results against the database that already exists. In comparison to other systems, face recognition
is typically more sophisticated and effective. The involved steps are listed below.
Fig: - Proposed System Architecture
3.1 Capturing Image from Video

The camera module can set in a district where individuals go into school or office and video is taken inside the distance
under 5 meters. A camera is utilized for taking video which contains many edges from which any of the casings can be
utilized for face acknowledgment and denoting the participation.
3.2 Detecting Face

Picking a productive calculation for face acknowledgment is basic in this proposed work. There are many face
recognition calculations accessible in OpenCV like Eigenfaces, Fisherfaces and Neighborhood Double Example
Histograms. Taking into account the requirement for the ongoing acknowledgment a calculation which has been picked
is the Haar Fountain Calculation/CNN [5] for face identification and acknowledgment. It is accessible in OpenCV
source library [6] and has ended up being strong [7].
3.3 Pre-processing
Since a picture might contain pointless foundation commotions and components other than faces eliminating those
elements is significant. Subsequently highlight extraction is key for lessening the picture to just a face accessible in the
picture. By this technique, the picture is decreased to a size of 150x150. Histogram adjustment is performed on the
diminished picture and in this way the picture becomes more straightforward to process.
3.4 Face Recognition and Classification

In this step the extracted feature for input image and trained model are been compared and accordingly the system
classify the result using CNN techniques .

www.ijarsct.co.in
IJARSCT

IV. CONCLUSION
Computerized Participation Framework has been imagined for the point of lessening the mistakes that happen inside the
conventional (manual) participation taking framework. The point is to computerize and make a framework that is
valuable to the association like an establishment. The effective and precise technique for participation inside the
workplace climate which will supplant the old manual strategies. This technique is adequately secure, dependable and
accessible to be utilized. No requirement for particular equipment for introducing the framework inside the workplace.
It are much of the time developed utilizing a camera and PC.
REFERENCES
[1]. W. Zhao, R. Chellappa, P. J. Phillips, and A. Rosenfeld,“Face recognition: A literature survey,” ACM
Computing Surveys, 2003, vol. 35, no. 4, pp. 399-458.
[2]. Herbert Bay, Andreas Ess, Tinne Tuytelaars, and Luc Van Gool. Surf: Speeded up robust features. Computer
Vision and Image Understanding (CVIU), 110(3):346–359.
[3]. H.K.Ekenel and R.Stiefelhagen,Analysis of local appearance based face recognition: Effects of feature
selection and feature normalization. In CVPR Biometrics Workshop, New York, USA, 2006
[4]. IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 4, No 1, July 2012 ISSN (Online): 1694-
0814
[5]. Javier Ruiz Del Solar, Rodrigo Verschae, and Mauricio Correa. Face recognition in unconstrained
environments: A comparative study. In ECCV Workshop on Faces in RealLife Images: Detection, Alignment,
and Recognition, Marseille, France, October 2008.
[6]. Kyungnam Kim “Face Recognition using Principle Component Analysis”, Department of Computer Science,
University of Maryland, College Park, MD 20742, USA.
[7]. Osuna, E., Freund, R. and Girosit, F. (1997). "Training support vector machines: an application to face
detection." 130-136

www.ijarsct.co.in

Real Time Web-Based Smart Attendance System Using AI

Uploaded by

Real Time Web-Based Smart Attendance System Using AI

Uploaded by

ISSN (Online) 2581-9429

Volume 3, Issue 2, April 2023