International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169

Volume: 5 Issue: 12 24 – 27
_______________________________________________________________________________________________
Automatic Attendance Using Face Recognition

Ms. Pooja Humbe, Ms. Shivani Kudale, Ms. Apurva Ms. Akshata Jagtap, Prof. Krushna Belerao
Kamshetty Trinity College of Engineering and Research
Trinity College of Engineering and Research University of Pune, India
University of Pune, India akshatajagtap36566@gmail.com
humbepooja888@gmail.com krushnabelerao@gmail.com
shivanikudale0@gmail.com
apurvakamshetty@gmail.com

Abstract— Human face detection and recognition is an important technology used in various applications such as video monitor system.
Traditional method for taking attendance is Roll Number of student and record the attendance in sheet which takes a lot of time. Because of that
systems like automatic attendance is used. To overcome the problems like wastage of time, incorrect attendance, the proposed system gives a
method like when he enters the class room , system marks the attendance by extracting the image using Principal Component Analysis
algorithm. The system will record the attendance of the student automatically. The student database is collected, it includes name of the students,
there images and roll number. It carries an entry in log report of every student of each subject and generates a pdf report of the attendance of the
student.

Keywords- face detection and recognition, Principal Component Analysis(PCA), MySQL, feature extraction.

__________________________________________________*****_________________________________________________

I. INTRODUCTION 1. Analog image processing

Student should attend the lectures regularly. This has at 2. Digital image processing
the end of a semester constituted part of student final grade Two dimensional analog signals is processed by analog
.To mark attendance manually which take a lot of time from image processing. Digital image processing is the use of
the teaching process. To create and maintain the record of computer algorithms or image processing on digital images.
attendance is compulsory in all schools, colleges. Hence Digital image processing is performed on a two dimensional
there are so many approaches of taking attendance like, image by a digital computer.
automatic attendance. Image processing is a special kind of Facial recognition technology is a new way of identify
signal processing which performs operations on images, to people. It works by picking faces out of a crowd, obtaining
extract useful data from it. Input is image, and output is the measurements and comparing it to the images already
image with great features. A Face Recognition system is an present in the database. The manual method will be replaced
application of image processing which performs two major by automatic attendance system , which is takes a lot of time
tasks of identifying and verifying a person from an image .In and is hard to maintain.
traditional attendance system there are some issues like fake In general, there are two known approaches to HFR,
attendance , time consumption, manipulation of attendance i.e. feature-based and brightness-based approach. The
and information cannot be secure. It is an application of feature- based approach uses key point features of the face,
identifying or verifying a person from a digital image. It is such as edges, eyes, nose, mouth, or other special
by comparing some facial features from the image and a characteristics. Therefore, the calculation process only
face database. Nowadays digital ease of use is very popular covers some parts of the given image that have been
than pen and paper.This digital growth results is popular for extracted previously. On the other hand, the brightness-
less time consuming and accurate user identification and based approach calculates all parts of the given image. It is
verification. So, there are number of algorithms are also known as holistic-based or image-based approach.
available from which we included Principal Component
Analysis (PCA) algorithm. Eigenface Approach
Eigenface approach is used ,it transforms faces into a
II. LITERATURE SURVEY set of characteristics, eigenfaces which is considered as
Image processing is consists of the input image, a training data. Recognition is done by projecting a new
photograph; the output of may be an image or a set of image in the eigenface subspace, in which person is
characteristics or parameters respect to the image. Image classified by comparing its position in eigenface space with
processing is classified into two types. They are, the position of known individuals.
24
IJRITCC | December 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 12 24 – 27
_______________________________________________________________________________________________
Set of eigenvectors are known as eigenvectors if they B. Hardware
are used for human face recognition. The covariance matrix Camera
of the probability distribution is used for derivation of
eigenvectors. To construct the covariance matrix, eigenface
form a basic set of all images. To represent the original
training images, it produces dimension reduction by
allowing the smaller set of basis images Classification can
be achieved by comparing how faces are represented by the
basis set.
1. In this paper , Initially video clip of classroom is
Specifications
taken and is stored in the database, then video is converted
Pan Angle: 360 degree
to images, then apply the face detection techniques to detect
Connectivity: Ethernet, Fireware Focus Range: 100-150cm
the faces and then features extraction is done by Histogram
It has Night Vision and Devices: Laptop, PC
of Oriented Gradients and Local Binary Pattern algorithm.
Video Sensor Resolution: 1280*720 pixel Still Image Sensor
[1]
Resolution: 720 MPHD
2. In this paper, a method for student attendance
system in classroom using face recognition technique by
IV. METHODOLOGY
using Discrete Wavelet Transforms and Discrete Cosine
The first step is that, the staff is provided with their
Transform to extract the features of student’s face which is
own Username and Password to Log-in. Next step is, the
followed by use of Radial Basis Function for classifying the
training image and their features are stored in the database.
facial components. [2]
Then, testing image features are compared with the training
3. In this paper, the proposed system describes a
images. Once the image is identified, the attendance will be
method like when he enters the class room and marks the
registered. Finally, the attendance details of the student are
attendance by extracting the image using Personal
send to staff and parent through E-Mail.
Component Analysis algorithm. The system will mark the
1. Login.
attendance of the student and it will maintain a log of each
2. Staff & Student Information.
student of each subject and also generates a pdf report of the
3. Update Attendance.
student attendance. Using Simple Mail Transfer Protocol the
4. Feature Extraction.
report will be sent to the faculty and also to the parents. [3]
5. Feature classification
III. SYSTEM REQUIREMENT
V. SYSTEM ARCHITECTURE
A. Software
PP Controller
XAMPP Controller is open source cross-platform, web
server solution database free software which contains the
MariaDB database Apache HTTP Server, and interpreters
for scripts which are written in the Perl and PHP
programming languages.

NetBeans
NetBeans is written in java,it is a software development
platform. Applications are developed from modules.
Applications are designed on the NetBeans Platform, with
integrated development environment (IDE), it can be
extended by third party and other developers.

Frontend
Java Advance(JSwing,Applet)

Backend
MySQL
Figure 2. System Architecture
25
IJRITCC | December 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 12 24 – 27
_______________________________________________________________________________________________
Image Acquisition 4. Compute the eigenvalues and eigenvectors of
Images can be acquired by facial-scan technology from the covariance matrix.
camera that captures images of better quality and resolution. 5. Choosing component and forming a feature vector.
6. Deriving new data coordinates.
Pre-processing 7. Approximation.
First Image is cropped from acquired image. Then
cropped images are resized to some pixels for face VII. CONCLUSION
recognition. These resized images are converted from RGB The automated student attendance system using human
to Gray level. face recognition technique works nicely.The automatic
attendance management will replace the traditional method,
Database (DB) which takes a lot of time and is hard to maintain.
It stores the pre-processed images for further processing Certainly, it is improved for better result particularly by
and results. paying attention in feature extraction or recognition process.
This improvement may help the recognition process become
more robust.
Template Matching
It compares match templates against enrollment
VIII. FUTURESCOPE
templates. In identifying a single individual from a large
The current developed software is installed on the system,
database, facial scan is not so effective as iris scan. After
i.e. it is a desktop application, and it will be used for some
large-scale facial-scan identification searches, numbers of
institute. But later it can be updated so that it will be operate
matches are returned.
as online application. Currently, the system has reached up
to some great accuracy level for partial and dense images. It
Face Recognition
can further be improved to obtain higher accuracy level. It
For face recognition or detection it compares selected
can be automatically updated by the use of the concept of
facial components from the image and a face database, it
Internet of Things.
identifies or verifies a person in image.

ACKNOWLEDGMENT
Face Database Generation
The project on “Automatic attendance using face
Original face database consists of images of all students
recognition“ is a source of trending and immense
having 5 images per student. With change in intensity of
information to us. We would glad to express our sincere
light and various facial expressions, the original database
gratitude to the “Trinity College of Engineering and
images are acquired at various interval of time.
Research department of Computer Technology for the
guidance and most valuable support in the help for this
VI. ALGORITHM project work. We acknowledge with a pleasant gratitude, the
Principal Component Analysis (PCA) encouragement and inspiration received from our guide
In this , a face contains set of important feature and Prof.Krushana Belerao, Head of the Department
these are called Principal Components or Eigen Faces. By Prof.Dr.S.B.Chaudhari , Principal Dr. P. Dabeer and
implementing PCA Technique facial features are extracted colleagues.
from original database.
Usage of PCA algorithm
REFERENCES
1. For data analysis [1] Aziza Ahmedi , Dr Suvarna Nandyal.” An Automatic
2. In data mining and machine learning. Attendance System using Image Processing.”The
International Journal Of Engineering And Science (IJES) ||
Benefits of PCA Algorithm Volume || 4 || Issue || 11 || Pages || PP - 01-08|| 2015 ||ISSN
1. Very robust to outliers and data corruption (e): 2319 1813 ISSN (p): 2319 1805
2. Speed and performance is good. [2] Samuel Lukas , Aditya Rama Mitra , Ririn Ikana Desanti ,
3. Identifying moving objects Dion Krisnadi “Student Attendance System in Classroom
Using Face Recognition Technique.”
4. Facial recognition.
ieeexplore.ieee.org/document/7763360/
[3] D. Nithya.“Automated Class Attendance System based on
Steps Face Recognition using PCA Algorithm.”Avinashilingam
1. Reduction of data into single vector. Institute for Home Science and Higher Education for
2. Calculate mean of the data. Women, Coimbatore, India. (IJERT)ISSN: 2278-0181Vol.
3. Calculate the covariance matrix. 4 Issue 12, December-2015.
26
IJRITCC | December 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
 International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 12 24 – 27
_______________________________________________________________________________________________
[4] Dr. Nita Thakare , Meghna Shrivastava , Nidhi Kumari [19] S. Choubey, Y. S. Bute “Discrete wavelet transform for
,Neha Kumari , Darleen Kaur , Rinku Singh .”Face image processing,” International Journal of Emerging
Detection And Recognition For Automatic Attendance Technology and Advanced Engineering, vol. 4, issue 3,
System.” International Journal of Computer Science and March, 2015
Mobile Computing, Vol.5 Issue.4, [20] Sumayyah Redhwan Allaam, Bayan Ali Saad Al-
[5] Shireesha Chintalapati, M. V. Raghunadh,“ Automated Ghamdi. Human Face by Face Recognition System with
attendance management system using face 3D. Journal of Information & Communication Technology,
recognition algorithms.” (ICCIC), 2013 IEEE International 2010.
Conference, DOI: 10.1109/ICCIC.2013.6724266.
[6] Mangesh Owandkar.” Attendance Monitoring System using
Face Recognition.”International Research Journal of
Engineering and Technology (IRJET), www.irjet.net
.Volume: 04 Issue: 05 | May -2017
https://www.irjet.net/archives/V4/i5/IRJET-V4I5228.pdf.
[7] Abhishek Jha ,” Class Room Attendance System Using
Facial Recognition System.”The International Journal of
Mathematics, Science, Technology and Management (ISSN
: 2319-8125) Vol. 2 Issue 3
[8] E.Varadharajan,R.Dharani, S.Jeevitha, B.Kavinmathi,
S.Hemalatha,” AUTOMATIC ATTENDANCE
MANAGEMENT SYSTEM USING FACE
DETECTION”, ieeexplore.ieee.org/document/7916753/
[9] Jyotshana Kanti, Shubha Sharma,” Automated Attendance
using Face Recognition based on PCA with Artificial
Neural Network.” Volume 3 Issue 6, June 2014
www.ijsr.net., Paper ID: 0201420.
[10] Yohei KAWAGUCHI ,Tetsuo SHOJI, Weijane LIN,Koh
KAKUSHO, Michihiko MINOH,” Face Recognition based
Lecture Attendance
System.”https://www.researchgate.net/publication/2416086
17_Face_Rec ognition-based_Lecture_Attendance_System.
[11] A. R. Mitra, R. I. Desanti., L. Samuel, D.
Krisndi.,“Implementing Discrete Wavelet and Cosine
Transform with Radial Basis Function Neural Network in
Facial Image Recognition”, Journal of Image and Graphics,
June 2016.
[12] Khayam S.A, The Discrete Cosine Transform
(DCT):Theory and Application, Michigan State University,
2003
[13] Nikunj Jain, Mr. Manish Kumar, Mayank Agarwal and
Himanshu Agrawal. Face Recognition with Eigen Faces,
Artificial Neural Network. International Journal of
Computer Engineering, August, 2010.
[14] U. B. Desai, V. V. Kohir, Face Recognition Using a DCT-
HMM Approach, Indian Institute of Technology,
Mumbai,India, 1998.
[15] R.W. Jasutkar, Y. S. Bute “Implement discrete wavelet
transform for image compression,” (IJCSN), 2012
[16] Z. Shaaban, M. Alwakeel “Face recognition based on
wavelet transform, principal component analysis with
Levenberg-Marquardt back propagation neural network,”
European Journal of Scientific Research, 2010.
[17] S. E. Handy, S. Lukas, “Further tests for face recognition
using discrete cosine transform with HMM,” in Proc.
(MICEEI), Makasar,2012.
[18] J.Nalini, A.A. Blessie and S.C.Ramesh, “Image
compression using wavelet transform based on the lifting
scheme and its implementation,” International Journal of
Computer Science Issues, 2011.
27
IJRITCC | December 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________