International Journal of Innovative Technology and Exploring Engineering (IJITEE)

ISSN: 2278-3075, Volume-8 Issue-4S2 March, 2019

Face Recognition Based Attendance System For

CMR College of Engineering and Technology

Kalachugari Rohini, Sivaskandha Sanagala, Ravella Venkata Rathnam, Ch.Rajakishore Babu 

Abstract: In modern times, Automatic Face recognition have implemented with different types of algorithms. In
(AFR) technologies have seen dramatic improvements in recent years CNN’s(convolution Neural Networks) have
performance over the last few years. There are two reasons for made significant breakthrough in an image classification.
this trend; the first is for saving the time in the classroom and
This face recognition systems have been attracting many
accuracy in attendance will be maintained, and the second is
availability of advanced technology it is more useful for the researchers to this side. To this research we also use the
future generation. In simple words, it was a computer deep learning model Deep ID developed by the research
implementation for recognizing automatically whether the group members only. This group have achieved the rate of
student is present in the classroom or not with the help of still 99.15% on IFW database, which is higher than that of
image or video frame. We proposed an automatic attendance human eyes with 97.52% By this system of developing an
management system. It was completely based on face
intelligent and automatic process may require human
recognition and the face detection. This both detection and
recognition will automatically detect the students in the computer interaction with this we can increase some of the
classroom and mark the attendance by recognizing the person. efficiency by uing the face recognition algorithms with the
This research includes for Face detection Students and system compare of neural network in this aspect (or )field.
is based on CNN perspectives and algorithms.
Keywords: Face Recognition, Face Detection, CNN, AFR, II. RELATED WORK
Deep Learning.
2.1. RFID (Radio Frequency identification) based on
I. INTRODUCTION attendance system:
Attendance is one of the important aspects in the In this RFID based system have been proposed where the
classroom. At the beginning and ending of the section, the students carry their ID card that card has the radio frequency
teacher should check out the students presence in the class waves with the help of that waves the reader will record the
generally teachers take the attendance and note done in the attendance when the card is placed on the reader .By this
register, but it may appear that miss someone and students system any authorized person may use and enter the
any give attendance multiple times.so to avoid this type of classroom or organization.
problems there is a high definition monitor video. Based on 2.2. CNN (Convolutional neural network):
this monitor we use face recognition and other information
In machine learning a Convolutional Neural Networks
technology aspects for marking the attendance, it is one of
(CNN’s or ConvNet)it is a class of deep learning. Feed-
the practical research technique.
forward network analyzing visual imagery where commonly
Firstly, we use CNN algorithm for face detection module applied. In this multilayer perceptron’s designed for the sake
to determine the number of students present in the class. of minimal preprocessing method ,It is known as SIANN
Recent years we observe that researchers on face
(Shift invariant or space invariance artificial neural network
recognition techniques have gained drastic change in with different fields of neurons the entire visual fields are
development. Compare will the biometric methods the covered by partially overlap.
fastest and accurate result will be progressed by the
automated attendance. For this by the automated attendance III. METHODOLOGY
that is face recognition technology identifies the facial parts
of a human face e.g. eyes, mouth, nose, ears, eyebrows etc, The attendance system based on face detection and face
by this parts of the face can recognize the particular student. recognition algorithm involved. Here are some of the
The second one is global face recognition uses complete modules as followed:
features to identifies the student so these two approaches • Image or video will be recorded to known that every
student is present in the image should be seen.
• Separate as per the detected images in the class
Revised Manuscript Received on March 02, 2019.
Kalachugari Rohini, M.Tech Scholar ,Department of CSE,CMR
attendance.
College of Engineering & Technology, Hyderabad, Telangana, India • Apply deep learning algorithm CNN for face detection
(Email: [email protected]) model detect the students faces as output.
sivaskandha sanagala, Associate professor Department of CSE,CMR
• Apply deep learning of face recognition algorithm for
College of Engineering& Technology, Hyderabad, Telangana, India
(Email: [email protected]) the sake of face recognition.
Ravella Venkata Rathnam, Assistant professor Department of
CSE,CMR College of Engineering& Technology, Hyderabad, Telangana,
India (Email: [email protected])
Ch.Rajakishore Babu, Associate professor Department of CSE,CMR
College of Engineering& Technology, Hyderabad, Telangana, India

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: D1S0026028419/19©BEIESP 127 & Sciences Publication
 Face Recognition Based Attendance System For Cmr College Of Engineering And Technology

• In this module automatic attendance,The students We observed algorithm gives the better output in different
seating order should be fixed, so that there will not be conditions to combine the multiple classifiers for better
any contrast their faces. The device can easily identify detection rate. Up to 30degrees angle the image is going to
the faces. be detected.
4.3. Database Development:
In every individual requirement biometric based system
enrolment can be chosen. In this phase we consider the image
capture to way person as individual features. We extract the
face and store the features in the database after identifying.
4.4. Proposed Algorithm:
1. The person’s image will be captured.
2. Apply detection algorithms for detecting the face.
3. Face recognition is going to be recognized.
4. Apply pre-processing image.
5. If enrollment phase
Then store in database
Else
Apply feature extraction
Apply for classification
End if
6. Post-processing.

V. ATTENDANCE RESULTS
“A new method of face recognition using convolutional Table.1 Attendance Statistics of Six Students
neural network”, Patrik KAMENCAY,Miroslav BENCO, Student Time(min)
Tomas MIZDOS, Roman RADI. 1 2 3 4 5 6 7 8 9 10
Tanmai         × ×
IV. PROPOSED WORK Shruthi          
System design some of the interfaces, components, Sandeep          
architecture and database to satisfy the requirements. There Rohith          
are some of system design applications. Proposed automated Mahanth         × ×
attendance system can be divided into few modules. Bharath          
4.1. Image capture: Finally, The attendance have sample every minute for ten-
Camera will capture the image of the students at the minute class and given the attendance statistics for six
beginning after capturing the image the next process is to students in a table format which we can see(1) Rohith
detect the image it goes to the face detection. achieves a full attendance, basically the students has facing
the blackboard correctly;(2) Tanmai and mahanth has the
4.2. Face Detection: problem with the early departure, one minute before
The face detection algorithm will increase the efficiency departure mahanth have appeared carelessness(3) Shruthi,
of the face recognition. There are some of the algorithms Sandeep, Bharath appeared carelessness, Where Shruthi is in
was proposed for face detection as face geometry-based the first 3-5 minutes didn’t listen the class carefully,
performances. Sandeep is in the first 4-6 minutes and Bharath only in the
sixth minute didn’t listen class carefully.

Image Face Pre Data VI. CONCLUSION

capture detection processing base of
faces In the system automatic attendance have been
implemented. By using the techniques proved time saving
and high-security. It also identify an unknown person too.
Feature
extraction The future extraction is to increase the rate of recognition
Feature
extraction
techniques. When there is a change in a student’s face, like
covering the head with the beard. This system has only
Classification developed face recognizes up-to 30 degree’s angle should be
improved further. The class room with a poor lighting may
effect image with lack of clarity system performance
degrades indirectly.
Name of recognized
person

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: D1S0026028419/19©BEIESP 128 & Sciences Publication
 International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-4S2 March, 2019

REFERENCE
1. R.Girshick,”Fast R-CNN,” in IEEE International
Conference on Computer Vision (ICCV), 2015.
2. Michal Dolecki, Pawel Karczmarek, Adam Kiersztyn,
Witold Pedrycz , “Face recognition by humans
performed on basis of linguistic and neural network”,
Neural Networks (IJCNN)2016 International Joint
Conference on , pp.5135-5140,2016,ISSN 2161-4407.
3. Soniya, Paul S, Singh L.A review on advances in deep
learning[c] IEEE Workshop on Computational
Intelgence: Theories, Applications and future Directions,
IEEE, 2015.
4. Xiaofei He; Shuicheng Yan; Yuxiao Hu;Niyogi,p.;
Hong-Jiang Zhang,IEEE Transactions on pattern analysis
and Machine Intelligence,pp.328340,2005.
5. H.Lu,K.N.Plataniotis, and A.N.Venetsanopoulos, Mpca:
Multilinear principal component analysis of tensor
objects, IEEE Trans. On Neural Networks,
19(1):1839,2008.
6. Ming Du, Aswin C.Sankaranarayanan, and Rama
Chellappa , Robust Face Recognition from Multi-View
Videos, IEEE “Transactions on Image Processing”,
Vol.23, No. 3,March 2014.
7. M.Turk and A.Pentland, Eigenfaces for recognition,
Journal of Cognitive Neuroscience, 3(1), pp.7186,1991.
8. Shuo Yang; Ping Luo; Chen Change Loy; Xiaous
Tang,”WIDERD FACE: A Face Detection Benchmark,”
2016 IEEE Conference on Computer Vision and Pattern
Recognition (CVPR) Year:2016 Pages:5525-
5533,DOI:10.1109/CVPR.2016.596.
9. Yang B,Yan J,Lei Z, et al.convolutional channel features
for pedestrian,Face and Edge Detection[J].
10. A.Krizhevsky, I.Sutskever and G.E.Hinton,” Imagenet
classification with deep convolutional neural network”,
in Advances in neural information processing system
25:26th Annual Conference on neural information
processing systems 2012.proceessing of a meeting held
on December 3-6,2012,(Lake Tahoe, Nevada, United
States),pp.1106-1114,2012.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: D1S0026028419/19©BEIESP 129 & Sciences Publication