Face recognition based

attendance system using

image processing

TEAM MEMBERS:
 RAGINI NANDAN SHAKYA(73)
UNDER THE GUIDANCE OF:
MR. SHAHID EQBAL  SALONI SINGH(80)
 RITIK PRATAP SINGH(75)
 Contents

• Introduction
• Brief Literature Review
• Motivation
• Problem Identification/Objectives
• System model/ block diagram/ flowchart/
methodology
• Plan of Actions
• References
 1

Introduction
Problems with Manual Attendance System
 Integration issues: Your time and attendance system does not adequately integrate with
your payroll program, requiring duplicate data entry.
 Multiple locations: You have multiple locations from which you need to collect data into a
central system.
 Lack of Flexibility: Your time and attendance system only performs basic functions, but
does not allow any customization or conformity.
 Unrealistic expectations: You are used to using manual times sheets and your goals for
automating your current process may not be in line with what is currently available on the
market.
 Budding punching: This is a common concern with employers- employees clocking in and
out for each other. This is often referred to as time theft.
 2
What is Fingerprint scanning???

 Fingerprint scanning is a biometric method ,because it involves the automated

capture,analysis and comparison of a specific characteristics of human body.
 Fingerprint scanning is the process of electronically obtaining and storing human
fingerprints. The digital image obtained by such scanning is called a finger image.
 Fingerprint scanner can bring out the details in the pattern of raised areas (called
ridges) and branches(called bifurications)in a human finger image.
 Biometric fingerscanning offers improvements over ink and paper imaging.
BLOCK DIAGRAM OF FINGERPRINT 3

SCANNING
 Enrollment module : An automated
mechanism that scans and captures a digital or
an analog image of a living personal
characteristics

 Database : Another entity which handles

compression, processing, storage and compression
of the captured data with stored data

 Identification module : The third interfaces with

the application system
 4
Why we choose face recognition
over other biometric?
There are number reasons to choose face recognition.
This includes the following :
 It requires no physical interaction on behalf of the user.
 It is accurate and allows for high enrolment and verification rates.
 It does not require an expert to interpret the comparison result.
 It can use your existing hardware infrastructure, existing cameras and image
capture. Devices will work with no problems
 It is the only biometric that allow you to perform passive identification in a one to.
 Many environments (e.g.: identifying a terrorist in a busy Airport terminal
Technologies Used

 Python 3
 D-lib Library
 Machine Learning
 Deep Learning
 Anaconda Tensor Flow
 JSP and Servlet
 MySQL
 Android Studio
S NO
Brief Literature Review
RESEARCH PAPERS PARAMETERS EVALUATED OBSERVATION

1
A. K. Jain, Y. Chen, andM. Demirkus, AUTOMATIC fingerprint recognition The experimental results show that our
“Pores and ridges: High-resolution systems (AFRS) are currently one of pore extraction
fingerprint matching using level 3 the most widely used biometric systems. method performs better than the state of
features,” IEEE Trans. Pattern Anal. fingerprint scanning, sweat pores, one of the art methods.
Mach. Intell., vol. 29, no. 1, pp. 15–27, the Level 3 features of fingerprint,
Jan. 2007. have been successfully used in
automatic fingerprint recognition
systems (AFRS). Since the pore extraction
process is a critical step for AFRS.
John M. Shrein (2017), “Fingerprint The proposed
2 classification using Convolution
The uniqueness of fingerprint is generally approach has achieved 98.21%
Neural Networks and Ridge classification accuracy with 0.9
determined
Orientation Images”, IEEE loss. The obtained accuracy is
by ridge features as well as minutiae
international significantly higher than
features, such as bifurcation and spots
conference, 978-1-5386-2726-6/17. previously reported results on the same
Craig I. Watson. (1992). NIST Special (ridge endings).
dataset, 77%.
Database 4:8-bit Gray Scale
Images of Fingerprint Image Groups,
National Institute of standards
and Technology. Experimental results show that our
A. Almansa and T. Lindeberg, Fingerprint recognition has
3 proposed algorithm is able to handle
“Fingerprint enhancement by shape emerged as one of the most reliable various input image contexts and
adaptation of scale-space operators means of biometric authentication achieves better results compared
with automatic scale-selection,” because of its universality, distinctiveness, with some state-of-the-art algorithms over
IEEE Trans. Image Process., vol. 9, no. permanence, and accuracy. public databases, and to
12, pp. 2027–2042, Dec. improve the performances of fingerprint-
2000. authentication systems.
 Motivation
 This project is being carried out due to the concerns that have been highlighted on
the methods which lectures use to take attendance during lectures.

 The use of clickers, ID cards swiping and manually writing down names on a sheet of
paper as a method to track student attendants has prompted this project to be
carried out.

 This is not in any way to criticize the various methods used for student attendance, but
to build a system that will detect the number of faces present in a classroom as well as
recognizing them.

 Provides an automated attendance system that is practical, reliable and eliminate

disturbance and time loss of traditional attendance systems.

 Present a system that can accurately evaluate student’s performance depending on

their recorded attendance rate.
 Objective
 Provides an automated attendance system that is practical, reliable and eliminate
disturbance and time loss of traditional attendance systems.
 Present a system that can accurately evaluate student’s performance depending on their
recorded attendance rate.
 Provides a valuable attendance service for both teachers and students.
 Reduce manual process errors by provide automated and a reliable attendance system
uses face recognition technology.
 Increase privacy and security which student cannot presenting himself or his friend while
they are not.
 Produce monthly reports for lecturers.
 Flexibility, Lectures capability of editing attendance records.
 Calculate absenteeism percentage and send reminder messages to students.
System Algorithm
• This section describes the
software algorithm for the
system.
• The algorithm consists of the
following steps:
 Image acquisition
 Histogram normalization
 Noise removal
 Skin classification
 Face detection
 Face recognition
 Attendance
Plan of Action
PHASE ACTIVITY DURATION
1 Project Proposal March,2019

2 Collect the database of all students July,2019

3 Implementation of face detection October,2019

4 Implementation of Predefined & Substituted December,2019

Algorithm

5 Implementation of face recognition February,2020

6 Maintain the attendance record April,2020

 References
 Carro, R. C., Larios, J. -. A., Huerta, E. B., Caporal, R. M. and Cruz, F. R. (2015) 'Face recognition
using SURF', Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial
Intelligence and Lecture Notes in Bioinformatics), 9225 pp. 316-326.
 Castrillón, M., Déniz, O., Hernández, D. and Lorenzo, J. (2011) 'A comparison of face and facial
feature detectors based on the Viola–Jones general object detection framework', Machine
Vision and Applications, 22 (3), pp. 481-494.
 Face Recognition', Applied Mechanics and Materials, 490-491 pp. 1338-1341.
 Da Costa, Daniel M. M., Peres, S. M., Lima, C. A. M. and Mustaro, P. (2015) Face recognition
using Support Vector Machine and multiscale directional image representation methods: A
comparative study. Killarney,Ireland. Neural Networks (IJCNN), 2015 International Joint
Conference on: IEEE.
 Dagher, I., Hassanieh, J. and Younes, A. (2013) Face recognition using voting technique for the
Gabor and LDP features. Dallas, TX, USA. Neural Networks (IJCNN), The 2013 International Joint
Conference on: IEEE.
 Dhanaseely, A. J., Himavathi, S. and Srinivasan, E. (2012) 'Performance comparison of cascade
and feed forward neural network for face recognition system', pp. 21.
Thank you