Mini Project

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 20

FACIAL RECOGNITION BASED

ATTENDANCE SYSTEM

Guided By: Submitted By:


Prof. Abhay E. Patil Pranav Ghasghase (415)
Murtaza Burhani (439)
Rohan Pophale (452)
Harish Yadav (477)
INDEX
 Abstract
 Introduction
 Aim & Objectives
 Literature Survey
 Advantages and Disadvantages of the Project
 Methodology
 Execution
 Future Scope of the Project
 Conclusion
 References
ABSTRACT
 Taking attendance is an important step to monitor the activities of a
student and to ensure the eligibility of the student to complete the course.
 Despite technological advancements, most of the educational institutes
still use the old register system.
 In this project, we propose a new way to take attendance of students in a
classroom using facial recognition technique
 This is efficient, less time consuming and which can be done using
devices that are readily available with people in today’s day and age.
INTRODUCTION
 Face recognition has been a very trending and developing area of
research.
 There is continuous research in the domain of face recognition and it has
reached a point where a computer recognises human face better than
humans ourselves.
 As such it has become a very reliable technique to use in security
applications like Theft monitoring systems, criminal database etc.
 Face recognition can also be used in educational institutes for taking
attendance. Manually taking attendance by calling out individual
student’s name one by name is a time-consuming process, hence
inefficient.
 Face recognition based attendance system is by far the best solution to
address this problem.
AIM & OBJECTIVES
AIM of the project:
• To assist in the process of attendance marking with the use of facial
recognition technique and record the attendance in a file.
• If there are multiple faces in a single picture, which is the case in
classrooms, the proposed model segments the entire image and fetches
individual faces separately.

OBJECTIVES of the project:


• To use Python and its libraries to design an application for facial
recognition.
LITERATURE SURVEY
Paper Name: Smart Attendance System using Facial Recognition
Authors: Harish M, Chetan P, Prajna N Holla K, Syed Abdul Azeem, M.G. Veena

A Fingerprint based attendance An RFID based attendance system Advantages of the Paper:
system was proposed by Sanjay proposed by T.S. Lim, S.C. Sim Use of biometric scanning way is
Badhe et al. which uses a and M.M. Mansor, uses a Radio definitely faster and better than the
Fingerprint module to take the Frequency Identification Device traditional one, as biometrics can
fingerprint of the student. This data (RFID) tag, which is unique to uniquely identify a person.
is wirelessly transmitted to a every individual. RFID cards are
microcontroller via a Zigbee placed on the reader and the Disadvantages of the paper:
wireless communication module. attendance of the student/worker However, the biometrics of a person
The fingerprint data is stored in a is instantly taken. can easily be forged. Hence, this
computer wherein the Fingerprint A Speech Biometric based system is not very secure.
is recognised based on the data attendance system proposed by
obtained from the Microcontroller. SubhadeepDey et al. In that user Research Gap:
accesses the system by making a This paper only uses biometric
call from few pre-decided mobile scanning as compared to the facial
phones. And this whole process is recognition technique being used in
conducted by an interactive voice our project.
response (IVR) system
LITERATURE SURVEY
Paper Name: Smart Attendance Management System Using Face Recognition
Authors: Kaneez Laila Bhatti, Laraib Mughal, Faheem Yar Khuhawar, Sheeraz Ahmed Memon

In Jomon Joseph, K. P. Zacharia A 3D face recognition approach for Advantages of the Paper:
proposed a system using image attendance management system was Facial recognition is a very fast and
processing, PCA, Eigen faces, proposed by MuthuKalyani.K, secure method of identifying a
Microcontroller, based on VeeraMuthu. There is need for an person.Facial recognition of a person
Matlab. Their system works only alternative algorithm which can enhance is very difficult to be forged. Hence,
with front face images and there the recognition on oriented faces. this method is generally foolproof.
is need of a suitable method Efficient Attendance Management system
which works with the orientation is designed with the help of PCA Disadvantages of the Paper:
of the system. algorithm the have achieved accuracy up The PCA model is 83% accurate, so
to 83%. not completely reliable.There are
Ajinkya Patil with their fellows
An eigen face approach along with PCA chances of students not being able to
in proposed a face recognition
algorithm for marking face recognition identify themselves under different
approach for attendance marking
attendance system have introduced by lighting conditions.
using Viola jones algorithm,
author in that they mention comparison of
Haar cascades are used to
different face recognition algorithm in Research Gap:
detect faces in images and
their paper. Overall it was good approach This paper uses the PCA model to
recognition performs through
to maintain record of attendance. identify faces, whereas our project
Eigen face method.
uses Image Processing to do the same.
LITERATURE SURVEY
Paper Name: Automatic Students Attendance Marking System Using Image Processing
Authors: Kaneez Laila Bhatti, Laraib Mughal, Faheem Yar Khuhawar, Sheeraz Ahmed Memon
There are so many recent works recognition rate as persons to be Advantages of the Paper:
done by many researchers in the identified increases. The methods illustrated in the
area of ‘Automatic attendance Dmello et al. proposed an paper increase the accuracy of the
marking systems’. attendance management system face recognition technique and
Sovitkar and Kawathekar used the based on detection and hence increase the reliability of
Viola Jones algorithm for recognition of student’s faces the system.
detection of the faces. For feature present in a class and mark Disadvantages of the Paper:
selection, they used algorithms recognized students’ attendance There are multiple algorithms
such as LDA, PCA and hybrid automatically. They used IOT used in a single system, which
approach using both of these with cameras to capture the students’ makes the programming part
SVM as a classifier. They created faces. CNN, along with SVM complex and difficult to debug.
dataset using different facial classifier is used for face Research Gap:
expressions, poses and lighting recognition. The system gives an The above paper uses complicated
conditions. Average rate of accuracy of 94%. algorithms and expensive
recognition with the combination hardware equipments, whereas
of all algorithms showed is 95%. our project uses simple devices.
There is decrease in the
ADVANTAGES AND DISADVANTAGES
OF THE PROJECT
Advantages of the Project:
• Fast and secure attendance marking system
• More accurate than the traditional methods
• Easy to install and implement

Disadvantages of the Project:


• The accuracy of the system depends upon the quality of hardware used.
• Problems with the angle or lighting conditions may interfere with the recognition
of faces.
• Some faces may go recognized if there are too many students in a single photo.
SYSTEM REQUIREMENTS
HARDWARE REQUIRED: SOFTWARE REQUIRED:
• A smartphone with good quality • Python 3.0 or higher
camera • PyQT
• Computer system with Windows • OpenCV
7 or higher • Cmake
PROPOSED MODEL
 The proposed model is an attempt to overcome all the shortcomings of the previous
related works in the domain of smart attendance system.
 If there are multiple faces in a single picture, which is the case in classrooms, this
method segments the entire image and fetches individual faces separately and
later it is sent through the Facenet model to get individual predictions.
 Images of the classroom is taken by the lecturer’s smartphone. The number of
images depends on the size of the classroom and the field of view of the smartphone
camera so as to capture the entire length and breadth of the classroom and not
miss out any student’s face.
 One can take multiple images of different sections of the classroom to avoid this
Field of view problem.
METHODOLOGY
Face recognition is essentially pattern recognition, and the purpose is to abstract
real things into numbers that computers can understand. If a picture is a 256 bit-
color image, then each pixel of the image is a value between 0 and 255, so we
can convert an image into a matrix.

How to identify the patterns in this matrix? One way is to use a relatively small
matrix to sweep from left to right and top to bottom in this large matrix. Within
each small matrix block, we can count the number of occurrences of each color
from 0 to 255. So we can express the characteristics of this block.
METHODOLOGY
• Facial feature extraction is a face image into a string of fixed-length numerical
process. This string of numbers is called the "Face Feature" and has the ability
to characterize this face.
• Human face to mention the characteristics of the process of input is "a face map"
and "facial features key points coordinates", the output is the corresponding
face of a numerical string (feature).
• Face to face feature algorithm will be based on facial features of the key point
coordinates of the human face pre-determined mode, and then calculate the
features.
• Earlier face feature models were larger, slow, only used in the background
service. However, some recent studies can optimize the model size and
operation speed to be available to the mobile terminal under the premise of the
basic guarantee algorithm effect.
FACIAL DETECTION FLOWCHART
EXECUTION
EXECUTION
APPLICATIONS AND FUTURE SCOPE OF THE
PROJECT
APPLICATIONS of the Project:
• The project can be deployed in various institutes and offices for facial
recognition based attendance management of students and/or staff.
• The project has applications in the security sector to only allow access to people
whose faces get a match in the dataset.
FUTURE SCOPE of the Project:
• The project can be deployed with a larger dataset with photos of a single
subject from various angles to improve the accuracy of the project.
• Various Machine Learning models can be deployed for better results in facial
recognition of the subjects.
• The dataset for the project can be rendered on a cloud so as to facilitate the
easy addition of more images to the dataset and referral of data for the Machine
Learning model.
CONCLUSION
• The proposed model for taking attendance is effective in reducing extra
hardware components required to take attendance.
• All the devices required for this purpose are already available with almost
everyone in the present day and age.
• This project is aimed to produce accuracy in facial recognition and attendance
marking.
• The accuracy of the model is dependent on the resolution of the camera with
which the image of the classroom is taken.
REFERENCES
• International Journal of Computer Applications National Conference on
Advancements in Computer & Information Technology (NCACIT), pp. 10-14,
2016.
• “Smart Attendance Monitoring System (SAMS): A Face Recognition based
Attendance System for Classroom Environment”, IEEE 18th International
Conference on Advanced Learning Technologies, pp. 358-360, 2018.
• “Face Recognition based Attendance Management System”, International
Research Journal of Engineering and Technology (IRJET), Volume: 05 Issue: 06,
pp. 1979-1985, 2018.
• "Implementation of classroom attendance system based on face recognition in
class." International Journal of Advances in Engineering & Technology 7.3
(2014): 974.
THANK YOU!

You might also like