Scribd
Upload
86 views15 pages

Melsreeieee

This document describes a proposed face detection-based student attendance registration system. It would work by storing face patterns for each student during registration and then using face recognition to match detected faces to the database for attendance marking. Teachers would simply press a button to start the attendance process. The system is linked to a timetable to know what class is occurring. It aims to automate attendance taking and prevent incorrect markings.

Uploaded by

melvin
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
86 views15 pages

Melsreeieee

This document describes a proposed face detection-based student attendance registration system. It would work by storing face patterns for each student during registration and then using face recognition to match detected faces to the database for attendance marking. Teachers would simply press a button to start the attendance process. The system is linked to a timetable to know what class is occurring. It aims to automate attendance taking and prevent incorrect markings.

Uploaded by

melvin
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

ATTENDANCE REGISTRATION

USING FACE DETECTION

Group Members
Guided by- Divya K.S
MELVIN SEBASTIAN
SREEHARI MOHAN
Department of INFORMATION TECHNOLOGY
INTRODUCTION
●
Patterns of face pictures of an single student is stored in
the databank at time of registration
●
The faces are then recognised from that of the image
then the procedure matches with the database.
●
The attendance marked on server if any of the face
identified from the database and it can be accessed by
everyone for numerous reasons.
●
The system is also attached with the time table unit that
robotically gets class at what time and what subject.
●
Teachers after coming to the class when taking
attendance they just press the attendance button to start
the attendance procedure.

Title of the Project 2


LITERATURE SURVEY
NO. YEAR JOURNAL AND AUTHOR DETAILS RESEARCH WORK

1 Dagher, I., Hassanieh, J. and Younes, A. (2013) Face recognition robust comparison method,
using voting technique for the the Chi Square Distance, was
Gabor and LDP features. Dallas, TX, USA. Neural Networks used as a matching
(IJCNN), The 2013 International algorithm. Four databases
Joint Conference on: IEEE. involving different Image
capturing conditions
2013 positioning, illumination
and expressions were used.
The best results were
obtained after applying a
voting technique to the
Gabor and the LDP features.

2 2012 Dhanaseely, A. J., Himavathi, S. and Srinivasan, E. (2012) Two different architectures
'Performance comparison of cascade are investigated and
and feed forward neural network for face recognition system', presented in this paper. The
cascade architecture
(CASNN) and feed forward
neural architecture (FFNN)
are investigated.

3 2014 Hiremath, P. S. and Hiremath, M. (2014) '3D Face Recognition Bio-metrics is used in
based on Radon Transform, PCA, computer science as a form
LDA using KNN and SVM', International Journal of Image, of identification and access
control. It is also used to
identify individuals in
Title of the Project groups that are under 3
surveillance.
PROPOSED METHODOLOGY
●
Tomaz algorithm is used for skin classification and for
face detection.
●
This method first calculates the pixel values then
subjected to normalization .
●
If the values are greater than 0.7 are identified as
skin and this method works quite well.
●
This method detects skin in test images but as every
method has errors and is prone to blur the image by
clothes and other disturbance.

Title of the Project 4


PROPOSED METHODOLOGY
● Patterns of face pictures of an single student is stored in the
Data bank at time of registration.
● The faces are then recognised from that of the image then
the procedure matches with the database.
● The attendance marked on server if any of the face identified
from the database and it can be accessed by everyone for
numerous reasons.
● This also habits the procedure that can be used for
attendance.
● The system is also attached with the time table unit that
robotically gets class at what time and what subject.
PROPOSED METHODOLOGY
● Teachers after coming to the class when taking attendance they
press the attendance button to start the attendance procedure.

● This method is highly protected method here no one can give the
attendance of the other and also saves a lot of time.

● The device is used to take picture that is attached to that of the


system takes the pictures endlessly to identify and also to find the
students those who are sitting in the classroom .
● For avoidance of false recognition of images skin classification
method is used.
● This skin classification method progresses the proficiency and the

correctness that acknowledgement Procedure.


PROPOSED METHODOLOGY
● Mainly it is considered all other images recaps the other images then kept as
black in skin classification method, it expressively improves truthfulness face
tracking procedure .
● Assembling of face images and the images that are mined geographies at
period procedure of registration is done by using of the face database .
● The algorithm that has been taken for is subdivided from system process.

Subsequent stages are comprised from the system algorithm.


1) Image Acquisition
2) Histogram Normalisation
3) Noise filtering
4)Classification of skin
5) Face tracking
6) Face identification and
7) Monitoring attendance
PROPOSED METHODOLOGY
ARCHITECTURE
plot
Image MT ImageSize Window SF TP FP FN Detection
Size Rate
Image1 1 Imresize(I) [20 20] 1.05 28 15 0 96.55
Image1 1 Imresize(I,0.75) [20 20] 1.05 28 9 0 96.55
Image1 1 Imresize(I,0.5) [20 20] 1.05 25 3 4 89.29
Image1 1 Imresize(I,0.25) [20 20] 1.05 7 0 22 24.13

Title of the Project 9


GRAPHS
Range Count True Count False
Positive Positive
0-9 0 0
40-49 7 0
50-59 35 7
60-69 77 16
70-79 69 16
80-89 20 0
90-99 10 0
Total 218 39

Range Count of True Positive and False Positives.

Title of the Project 10


RESULT AND DISCUSSION
● From the above illustrations, it can be seen that at a percentage range of
60-69, the system will perform at its best.
● Because the system in this report depends on transmission and
emission probability matrix, we use the Baum Welch Algorithm to
estimate these probability matrices during the training.
● The maximum log probability of the transition and emission
probabilities of the trained model is generated to obtain a random
sequence of states.
● With these sequence of states, the most likely paths through the model
specified by these probabilities is the probable state and calculated in
percentage.
● This explains why carrying out a different test will still obtain the same
results but the percentage range at which the system will perform best
will differ slightly. However, this will not be very far from a range of
60-69 or 70-79 as it has the highest cluster of recognition per image.
Title of the Project 11
REFERENCES
• 1.TolbaA S, El-BazA H and El-HarbyA 2014 Face Recognition: A Literature
Review International Journal of Signal Processing
• [2] SushmaJaiswal, Sarita Singh Bhadauria and Rakesh Singh Jadon
2015 comparison between face recognition algorithm Eigen faces fisher
faces and elastic bunch graph matching Journal of Global Research in
Computer Science
• [3] Ming-Hsuan Yang, David JKriegman and NarendraAhuja2002
Detecting Faces in Images Survey IEEE transactions on pattern analysis
and machine intelligence
• [4] Li X and Areibi S 2004 A Hardware or Software code sign approach
for Face Recognition The16th International Conference on Micro-
electronics Tunisia
• [5] Yue Ming, QiuqiRuan, Xiaoli Li and Meiru Mu 2014 Efficient Kernel
Discriminate Spectral Regression for 3D Face Recognition Proceedings
Of ICSP

Title of the Project 12


CONCLUSION
AND SCOPE
● The data required for the completion of the project has been
collected, arranged and the Model is being trained. The
model’s accuracy is being improved at the current stage for
more accurate results
SCOPE

● Face recognition and Attendance marking


● 2 or more IP cameras can be employed and each
image can be processed separately. The results of
these can be merged to obtain better results and
accuracy in denser classrooms.
Thank you

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy