Seminar Report On ATTENDANCE SYSTEM BASED ON FACE RECOGNITION USING EIGEN FACE AND PCA ALGORITHMS Submitted in partial fulllment of the requirements for the award of the degree of Bachleor of Technology in Computer Science

Seminar Report On ATTENDANCE SYSTEM BASED ON FACE RECOGNITION USING EIGEN FACE AND PCA ALGORITHMS Submitted in partial fulllment of the requirements for the award of the degree of Bachleor of Technology in Computer Science

Seminar Report On
ATTENDANCE SYSTEM BASED ON FACE
RECOGNITION USING EIGEN FACE AND PCA ALGORITHMS
Submitted in partial fulllment of the requirements for the award of the degree of
Bachleor of Technology in
Computer Science & Engineering
By
Sheenu George
(RET15CS171)
Under the guidance of Mr.Hareesh M J Department of Computer Science & Engineering
Ra jagiri School of Engineering and Technology
Ra jagiri Valley, Kakkanad, Kochi, 682039
December 2018

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
RAJAGIRI SCHOOL OF ENGINEERING AND TECHNOLOGY RAJAGIRI VALLEY, KAKKANAD, KOCHI, 682039 CERTIFICATE
Certied that the seminar work entitled Attendance system based on face recogni-
tion using Eigen face and PCA algorithms” is a bonade work done bySheenu
George University register number RET15CS171in partial fulllment of the award
of the Degree of Bachelor of Technology in Computer Science & Engineering from APJ
Kerala Technological University, Thiruvanathapuram, Kerala during the academic year
2018-2019.
Dr. Sminu Izudheen Mr. Paul Augustine Mr. Hareesh M J
Head of Department Project Coordinator Project Guide
Dept. of CSE Asst.Professor Asst.Professor
RSET Dept. of CSE Dept. of CSE RSET RSET

ACKNOWLEDGEMENT
I am thankful to my Head of the Department ,whose help and guidance has been a
ma jor factor in completing my journey.
Sheenu George

ABSTRACT
This topic represents an attendance system that uses face recognition technique.
Face recognition systems are built on the idea that each person has a particular face
structure, and using the facial symmetry, computerized face-matching is possible. The
images must be processed correctly for computer based face recognition. The face and
its structural properties should be identied carefully. In this paper, Eigen faces method
is used for face recognition. In the recognition process, an Eigen face is formed for the
given face image, and the Euclidean distances between this Eigen face and the previously
stored Eigen faces are calculated. The Eigen face with the smallest Euclidean distance is
the one the person resembles the most.

Contents
Acknowledgement ii
Abstract iii
List of Figures vi
List of Tables vii
1 Problem statement 1
2 Introduction 2
3 Relevance of the topic 3
4 Literature survey 4
5 Proposed method 6 5.1 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1.1 Enrollment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1.2 Image acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5.1.3 Grayscale conversion . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5.1.4 Histogram normalization . . . . . . . . . . . . . . . . . . . . . . . . 7
5.1.5 Noise removal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5.1.6 Skin classication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1.7 Face Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1.8 Face Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1.9 Attendance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
6 Architecture 10
7 Flow chart: Eigen faces and PCA method 11
iv

8 Algorithm: Eigen faces and PCA method 12
9 Flow chart: Face recognition for attendance system 15
10 Algorithm: Face recognition for attendance system 16
11 Applications 18
12 Conclusion 19
13 Future Scope 20
References 21
v

List of Figures
5.1 Enrollment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.2 Grayscale conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5.3 Histogram Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5.4 Noise removal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
6.1 Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
7.1 Eigen faces and PCA method . . . . . . . . . . . . . . . . . . . . . . . . . 11
8.1 Training set of images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
8.2 Average image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
8.3 Eigen faces of the training set . . . . . . . . . . . . . . . . . . . . . . . . . 14
9.1 Face recognition for attendance system . . . . . . . . . . . . . . . . . . . . 15
10.1 Face recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
vi

List of Tables
4.1 Comparison of various algorithms for face recognition . . . . . . . . . . . . 5

Chapter 1
Problem statement
To implement an attendance system that recognizes face using Eigen face and PCA
method.
1

Chapter 2
Introduction
There are large number of students in each class, so it is very dicult time consuming
to take the attendance. In this system at rst the image is captured using a camera
device that is placed inside the classroom. This image is given as the input to the system.
Then the image is enhanced using some image processing techniques such as grayscale
conversion. This enhanced image is given for face detection. After this process, the next
step is to recognize the face using Eigen face technique along with PCA. In Eigen face,
when faces are detected, they are cropped from the image various features like eyes , nose
etc. are extracted from it. Using these features the students are recognized by comparing
these features with the face database their attendance is marked.
2

Chapter 3
Relevance of the topic
Maintenance and monitoring of attendance records plays a vital role in the analysis
of performance of any organization. The purpose of developing attendance system is to
computerize the traditional way of taking attendance. This attendance system performs
the daily activities of attendance marking and analysis with reduced human intervention.
3

Chapter 4
Literature survey
From the literature survey done for the face recognition, it is a two step process with
face detection and face recognition.
In the face detection, the image area is classied into face and non face.
Dierent approaches for face detection are holistic approach,feature based approach and
appearance based approach.
In Holistic approach, the entire face region is considered as input to the face detec-
tion system.
In feature based approach, the features of face such as nose and eyes are segmented
and then taken as input in face detection system.
In appearance based approach, the entire facial appearance is considered as input
to the face detection system.
Eigen face technique is very sensitive to the head orientation. So the camera and the
scanner should be implemented for more
exibility.
4

Method No. of images Success rate Reference No.
Principal
Component
Analysis
(PCA) 400 79.65% 5
Principal
Component
Analysis +
Relevant
Component
Analysis 400 92.34% 5
Independent
Component
Analysis 40 Gauss
function 81.35% 8
Support
Vector
Machines – 85-92.1% 10
Neural
Networks – 93.7% 11
Eigenfaces
Method 70 92-100% 12
Eigenfaces
with PCA
Method – 92.30% 13
Table 4.1: Comparison of various algorithms for face recognition
5

Chapter 5
Proposed method
The basis of the Eigen faces method is the Principal Component Analysis (PCA). The
Principal Component Analysis is a method of pro jection to a subspace and is widely used
in pattern recognition. An ob jective of PCA is the replacement of correlated vectors of
large dimensions with the uncorrelated vectors of smaller dimensions.
The strategy of the Eigen faces method consists of extracting the characteristic features
on the face and representing the face as a linear combination of the so called Eigen faces
obtained from the feature extraction process.
5.1 Methodology
5.1.1 Enrollment Figure 5.1: Enrollment
Students are enrolled to the database using their faces and basic information. Enroll-
ment includes:
Taking image by camera
6

Enhancement of that image
Feature extraction
Maintain database
5.1.2 Image acquisition In front of the classroom a high denition camera is installed. This camera will capture
the image of the classroom. Then this image is given as input to the system.
5.1.3 Grayscale conversion Images captured by the camera are colored images. These images are converted to
grayscale images for enhancement. Figure 5.2: Grayscale conversion
5.1.4 Histogram normalization It is a technique used for contrast enhancement. Using this technique, students who
are sitting at the back rows can be clearly seen and it will be easy to recognize them.
7

Figure 5.3: Histogram Normalization
5.1.5 Noise removal Figure 5.4: Noise removal
The captured image may contain noise which has to be ltered. Median ltering is
generally used for removing noise.
8

5.1.6 Skin classication
All the pixels are made black except those pixels which are closely related to the skin.
Those pixels become white.
5.1.7 Face Detection In face detection, faces are cropped from the image. Features of the face like eyes,eyebrows,nose
etc are detected.
5.1.8 Face Recognition Face recognition is done using Eigen face and PCA method. Here we compare the
cropped faces with the enrolled faces in the face database.
5.1.9 Attendance Attendance is marked on the server after verication of faces and successful recognition.
9

Chapter 6
Architecture Figure 6.1: Architecture
10

Chapter 7
Flow chart: Eigen faces and PCA method Figure 7.1: Eigen faces and PCA method
11

Chapter 8
Algorithm: Eigen faces and PCA method
1. Start
2. Read training set of N * N images
3. Select training set of N * M where, M: number of sample images Figure 8.1: Training set of images
12

4. Find average face, subtract from the faces in the training set, create matrix A Figure 8.2: Average image
= M 1
P
M
i =1
i where
! Average image
i !
Image vector
i=

i

where i=1,2,3……M
A=
1;

2;

3::::

M
5. Calculate covariance matrix C: AA’
6. Calculate Eigen vectors of the c covariance matrix
7. Calculate Eigen faces
8. Create a reduced Eigen face space using PCA
9. Calculate Eigen face of image
13

Figure 8.3: Eigen faces of the training set
10. Calculate Euclidian distance between the Eigen face of image and the Eigen
faces of the training set
14

Chapter 9
Flow chart: Face recognition for attendance system Figure 9.1: Face recognition for attendance system
15

Chapter 10
Algorithm: Face recognition for attendance system Figure 10.1: Face recognition
1. Start
2. Enroll the students’ information in the face database
3. Install a camera device in classroom
4. Input the image taken by camera
5. Enhancement of image
1. Convert to grayscale image
2. Generate histogram of grayscale image
3. Equalize the image
16

4. Generate histogram of equalized image
5. Remove noise from image
6. Skin classication of image
6. Face Detection
1. Crop the faces of students form image
2. Select the region of interest
7. Face Recognition
1. Compare the cropped images with face database images
2. Mark the attendance on attendance server
3. If any other face, Then Go to II of step 6
8. End
17

Chapter 11
Applications
Security/Counterterrorism
Banking using ATM
Smart Cards and Face ID
Day Care
Residential Security
Attendance in companies
18

Chapter 12
Conclusion
This paper focus on an ecient attendance system for classroom attendance. Using
this system the chances of fake attendances and errors can be reduced. Many biometrics
systems are used for managing attendance, but the face recognition system has the best
performance. A camera device, a standalone PC and database servers are sucient for
constructing the smart attendance system.
19

Chapter 13
Future Scope
Future work can be done to alert the student by sending SMS regarding the attendance.
SMS alert can be given to the parent also. This can be implemented using GSM module.
20

References
1 Ajinkya Patil, Mridang Shukla, “Implementation of Class Room Attendance System Based on Face Recognition III Class”, IJAET (International Journal of Advances in
Engineering and Technology), Vol. 7, Issue 3, July 2014.
2 Nirmalya Kar, Mrinal Kanti Debbarma, Ashim Saha, Dwijen Rudra Pal, “Study of Implementing Automated Attendance System using Face Recognition Technique”,
IJCCE (International Journal of Computer and Communication Engineering), Vol.
1, No. 2, July 2012.
3 Abhishek Jha, “Class Room Attendance System Using Facial Recognition System”, IJMSTM, ISSN:2319-8125, Vol. 2, Issue 3.
4 Naveed Khan Baloch, M. HaroonYousaf, Wagar Ahmad, M. Iran Baig, “Algorithm for Ecient Attendance Management: Face Recognition based Approach”, IJCSI,
Vol. 9, Issue 4, No I, July 2012.
5 Yasaman Heydarzadeh, Abol Fazl ToroghiH aghighat, “An Ecient Face Detection Method using AdaBoost and Facial Parts”, IJSSST.
6 M. Gopi Krishna, A. Srinivasulu, “Face Detection System On AdaBoost Algorithm using Haar Classiers”, IJMER(International Journal of Modem Engineering Re-
search), Vol. 2, Issue 5, Sep-Oct 2012.
7 K. Susheel Kumar, Shitala Prasad, Vijay Bhaskar Semwal, R. C. Tripathi, “Real Time Face Recognition using AdaBoost Improved Fast PCA Algorithm”, IJAIA,
Vol.2, No. 3, July 2011.
8 Tiwari Priti Anilkumar, Kalyani Jha, Karishma P Uchil, Naveen H., “Haar Features Based Face Detection and Recognition for Advanced Classroom And Corporate At-
tendance”, IJIRCCE, Vol. 3, Issue 5, May 2015.
21

9 Ralph Gross, Vladimir Bra jovic, “An Image Preprocessing Algorithm for Illumi-
nation Invariant Face Recognition.”, International Conference on Audio and Video
Based Biometric Person Authentication, pp 10-18, June 9-11-2013.
10 0 Yongmin Li, Shaogang Gong and Heather Liddell “Support Vector Regression and Classication Based Multi-view Face Detection and Recognition”, University of
London.
11 1 Anjana Mall, Mrs. Shusmita Ghosh, “A Neural Network Based Face Detection Approach”, Int. 1.Computer Technology Applications, Vol 3 (2), 823-829, July 2010.
12 2 Ajit K. Ushir, Ajinkya A. Shete,Sanika N. Tiwari, Vasanti R. Vishwakarma, Mahesh Sanghavi, “Mathematical Modeling for Face Recognition”, International Conference
on Recent Trends in Engineering Technology(ICRTET).
13 3 A. M. Pati!, Dr. Satish R. Kolhe, Dr. Pradeep M. Pati!, “Face Recognition by PCA Technique”, Second International Conference on Emerging Trends in Engineering
Technology( ICETET)
22