ResearchPDF Available

Development of a Secure Access System

Authors:

Abstract and Figures

Facial identification from real data, photo capture, sensor images, and database images are challenging tasks because of the wide variety of face looks, illumination effects, and intricate image backdrops. Face recognition is one of the most practical and modern applications of image processing and biometric systems. We describe face recognition methods and algorithms that have been created by several researchers in the fields of image processing and pattern recognition utilizing HAAR and OpenCV in this paper. This essay will also discuss the face recognition system's use of this technology and how it differs from alternative methods in terms of effectiveness. Therefore, a general review of face detection studies and systems that rely on various methodologies and algorithms is included in this research. We implemented a complete face recognition system by integrating the best option for each step. With training and without training, it achieves superior performance on every category of the test. The tools which are required for completing the system are a Camera, Microphone, RFID, and Pin Generator. This research study also analyses the performance of various approaches and algorithms in addition to the strengths and weaknesses of these literature studies and systems.
Content may be subject to copyright.
ISSN: 2584-0495 Vol. 2, Issue 6, pp. 915-923
International Journal of Microsystems and IoT
ISSN: (Online) Journal homepage: https://www.ijmit.org
Development of a Secure Access System
Manjot Kaur Bhatia, Gurpreet Kaur, Rajat Tanwar, Jayant Marwaha, Mrinal
Narang
Cite as: Bhatia, M. K., Kaur, G., Tanwar, R., Marwaha, J., & Narang, M. (2024).
Development of a Secure Access System. International Journal of Microsystems and
IoT, 2(6), 915923. https://doi.org/10.5281/zenodo.13149061
© 2024 The Author(s). Published by Indian Society for VLSI Education, Ranchi, India
Published online: 24 June 2024
Submit your article to this journal:
Article views:
View related articles:
View Crossmark data:
DOI:
https://doi.org/10.5281/zenodo.13149061
Full Terms & Conditions of access and use can be found at
https://ijmit.org/mission.php
915
International Journal of Microsystems and IoT
Vol. 2, Issue 6, pp. 915-923 ; DOI: https://doi.org/10.5281/zenodo.13149061
Development of a Secure Access System
Manjot Kaur Bhatia, Gurpreet Kaur, Rajat Tanwar, Jayant Marwaha, Mrinal Narang
Department of Information Technology, Jagan Institute of Management Studies Delhi, India
KEYWORDS:
Face Recognition,Image Processing,
HAAR, OpenCV, Fingerprint
1. INTRODUCTION
Technology currently reduces the amount of work done by
humans; hence it is essential for ensuring ownership of an
individual's possessions. Instead of wasting time and money on
security, it was made possible by the automation of electronic
equipment. Bhatia, M.[2013, 2015, 2017] proposed secure user
authentication system by hiding password in images. This
project creates a system that thoroughly restrains human
substances by relying on technology. This technology can help
protect the privacy of any person or organization, such a bank
vault or a money locker, among others.
Considering a breach of security, the system is a
combination of biometric and prior tech to ensure security
escalation of this system. The objective is delivering security
progression over extremely concealed premises effortlessly at
an affordable cost. Surveillance and facial recognition being
operated by a camera for premium security purposes as it stores
surveillance data on a memory chip for a definite period. [7]
The camera is continuously searching for a human face, and an
authorizer is able to see the situation through the camera in real
time. To achieve this goal, wireless communication was
established between the security system and the authorizer via
VNC viewer. [9]
When a person places himself in front of the camera, the
camera immediately scans his face with the image database
stored in Raspberry Pi. The authorizer is able to see the name
of that person at the top of his face in real time video.
© 2024 The Author(s). Published by Indian Society for VL SI Education, Ranchi, India
If the face doesn't match with the database, the name will be
replaced by an unknown person, and the system generates a
sound which indicates the sound of an unknown person. After
facial recognition, the system will give you a choice for two
authentication methods it will ask for (Pin Password - RFID) or
(Pin Password - Voice Recognition) if the user fails to enter the
right password alarm will ring if the user enters the opposite
password. If a user enters a password that is the reverse of the
original password, the silent alarm will go to the police.
2. LITERATURE SURVEY
2.1 Bhattacharyya, Budhaditya, Akya Bhatnagar, and
Arya Bhattacharya. "A NOVEL APPROACH TO
AUTOMATED PARKING USING RFID BASED
USER AUTHRIZATION.” (2018).
This paper described that Law enforcement has effectively used
fingerprint matching for more than a century. These days, a
variety of uses for the technology are being made, including
identity management and access control. In this context,
research opportunities are presented together with a system for
automatically identifying fingerprints and spotting significant
issues. The description of an RTOS (Real-time operating
system) implementation in the context of an embedded system
in this report is written in a manner similar to that of a product
design. Even though it is widely used, fingerprint recognition is
a difficult pattern recognition problem.
Making accurate algorithms that can extract important
characteristics and robustly match them is a difficult task. [20]
ABSTRACT
Facial identification from real data, photo capture, sensor images, and database images are challenging
tasks because of the wide variety of face looks, illumination effects, and intricate image backdrops. Face
recognition is one of the most practical and modern applications of image processing and biometric
systems. We describe face recognition methods and algorithms that have been created by several
researchers in the fields of image processing and pattern recognition utilizing HAAR and OpenCV in
this paper. This essay will also discuss the face recognition system's use of this technology and how it
differs from alternative methods in terms of effectiveness. Therefore, a general review of face detection
studies and systems that rely on various methodologies and algorithms is included in this research. We
implemented a complete face recognition system by integrating the best option for each step. With
training and without training, it achieves superior performance on every category of the test. The tools
which are required for completing the system are a Camera, Microphone, RFID, and Pin Generator.
This research study also analyses the performance of various approaches and algorithms in addition to
the strengths and weaknesses of these literature studies and systems.
916
In this study, we provide a fresh approach to resolving current
problems using a suitable embedded system architecture.
2.2 Chin, Howard. Face recognition based automated
student attendance system. Diss. UTAR, 2018.
This paper proposed that the majority of institutions in
underdeveloped nations still track students' attendance using
paper sheets. The management of the students' attendance
records urgently calls for the adoption of an alternative
approach. RFID enables the university administration to use
cutting-edge new technology while taking into account factors
like dependability, time savings, and ease of control to improve
the university's monitoring system. This article describes the
design and development of a student attendance system in terms
of hardware and software.
The system is integrated with a database management system
to reach complete system capabilities, enabling real-time
information manipulation. The RFID Platform is used to
emulate RFID scanners. The RFID platform and the.NET
Framework have been used to create an automatic attendance
system that makes use of RFID.
2.3 Kasar, M., Bhattacharyya, D. and Kim, T. (2016). Face
Recognition Using Neural Network: A Review.
International Journal of Security and Its Applications,
10(3), pp.81-100
This paper described that due to the broad range of face looks,
illumination effects, and intricate image backgrounds, face
recognition from real data, capturing photos, sensor images, and
database images is a hard task. Face recognition is one of the
most practical and modern applications of image processing and
biometric systems. In this study, artificial neural networks
(ANNs), which have been used in the fields of image processing
and pattern identification, are used to cover the facial
recognition techniques published by various researchers. In
addition, this paper will explain how ANN will be used for the
face recognition system and why it is superior to other
approaches.
There are numerous ANN-suggested methods that offer an
overview of face recognition using ANN. This study presents a
review of the literature on ANN-based facial recognition
systems. In this paper, several face recognition system
architectures, approaches, algorithms, methodologies, databases
for training or testing image sets, and performance metrics are
examined. Every researcher has their own way of identifying
faces in databases or videos, and while many studies have
attempted to address the issues with the earlier suggested
method, there are still some benefits and drawbacks to the
methods we have mentioned.
2.4 Mikael Nilsson, Jorgen Nordberg, and Ingvar Claesson
“Face detection using local SMQT features and split up
SNOW classifier in IEEE International conference on
Acoustics, Speech, and signal processing
(ICASSP),2007, vol 2, pp. 589-592
This study serves two purposes. First, it proposes local
Successive Mean Quantization Transform features for object
recognition that operates in the absence of light and sensors.
Secondly, a split Sparse Network of Winnows is provided. The
classifier and attributes are integrated to fulfil the task of frontal
face detection. For the MIT+CMU and Bio ID databases, the
results of detection are displayed. The best-published result for
this face detector comes from the Receiver Operation
Characteristics curve for the Bio ID database. The outcomes of
the CMU+MIT database can be compared to modern face
detectors. A face detection system was developed by fusing the
local SMQT characteristics with the split-up classifier. The face
detector obtains the best reported ROC curve for the Bio ID
database and a ROC curve that is comparable to published,
cutting-edge face detectors for the CMU+MIT database.
2.5 Bifari, E.N.; Elrefaei, L.A. “Automated Fingerprint
Identification System Based on Weighted Feature
Points Matching Algorithm”, 978-1-4799-3080-
7114/$31.00 ©2014 IEEE
This paper shows that the majority of fingerprint identification
systems use matching algorithms based on various minute
fingerprint features. Typically, details are taken from the
fingerprint picture that has been thinned. The thinning image
might produce a lot of false minutiae as a result of the image
noise and various pre-processing techniques, which could hurt
the system's performance. In order to create the Automated
Fingerprint Identification System in this study, some pre-
existing algorithms from other studies were merged. On the
basis of a feature with two points, minutiae and ridge point, a
new matching algorithm was suggested. Additionally, a suitable
weight was applied to each extracted feature in accordance with
the recommended weights table. A database demonstrated its
effectiveness and showed that it outperforms the traditional
minutiae-based matching algorithm in terms of outcomes.
The procedures to develop AFIS via MATLAB were
described in this study along with some suggested changes to
the filter algorithm and a new matching algorithm. Five FVC
databases were used in the systems testing. Two points
minute and ridge pointswere combined as a fingerprint
characteristic, changing the minutiae-based matching process.
Then, a weight table was provided to provide a similarity value
based on the quality of each feature for each feature. The
experimental test demonstrates that the outcomes are
dependable and efficient. In order to achieve better results, we
advise integrating the suggested approach with other techniques
such as local minutiae matching in future studies.
2.6 Rowley, H.A., Baluja, S., Kanade, T. Neural Network-
based Face Detection. IEEE Trans. Pattern Anal.
Machine Intel, Vol. 20, No.1 Rowle
This paper presents that a fundamental issue in computer vision
is object detection. Simply being aware of an object's presence
or absence is helpful for applications like image indexing. Face
recognition and other technologies that often deal with people
depend on facial detection in particular. The object detection
problem can be solved using a variety of techniques, including
matching sets of two-dimensional images of the object and
matching two- and three-dimensional geometric models to
photographs.
This dissertation will show how artificial neural networks can
be utilised to successfully implement the later view-based
917
technique, which permits the detection of upright, slanted, and
non-frontal faces in crowded images. In this study, the majority
of the detector's training was done with real-world sample
photographs, both with and without faces. For this strategy, a
big training data collection is required.
3. PROPOSED MODEL
This model describes how our model performs different types
of work in a single system. The name of this model is
“Development of Secure Access Systematization”. In this
product, we can use it in many ways according to the
requirement of the client. As we can see in Fig. 1 the proposed
model of our system.
A system for managing the different requirements effectively.
It helps to overcome the situation now facing of using a manual
system like a signature, RFID, Finger Print based systems but
they fail to proof the system.
This product solves many problems which we faced during the
execution. This product removes the lack of security because
we provide the RFID which is very secure. Titan Security
Product also removes the time-consuming system. It provides
efficient storage for better data searching.
Fig.1 Proposed Model for Facial Recognition System
3.1 Working of device
Some steps would be followed strictly for working the device
successfully:
3.1.1 First person has to show his face to the camera for
Facial Authentication. If the face will be authenticated then
he/she will move on to the next level of authentication else a
security alarm will siren.
3.1.2 Camera will click photos of authenticated users and
non-authorized users with the date and time spontaneously at
the time of facial recognition and will send them to an
application/web portal which can be further reviewed by the
admin.
3.1.3 Then the person will have the choice to authenticate
two security layers from the rest of the three different security
layers (RFID, PIN, Voice Recognition).
3.1.4 If somebody fails to authenticate any of two layers
from the rest of the 3 security layers, then the buzzer will siren.
3.1.5 If a person uses a pin password as one of the
authentication methods from the rest of 3 and enters the
password reverse of the correct password a silent alarm will
inform the police about the threat.
3.1.6 After facial recognition, voice recognition would take
place where sample audio would be recorded of the person.
3.1.7 The sample audio would be processed to match the
phrase spoken by the person and along with that to match the
pitch of the person.
3.1.8 Once the process is passed by the system the door
would be opened.
3.1.9 Once the person has entered the premises a log file
would be updated where the IN time of the person would be
logged.
3.2 System Components
The HAAR-Cascade Detection in OpenCV provides the trainer
as well as the detector.
We can train the classifier for any object like cars, planes, and
buildings by using OpenCV.
There are two primary states of the cascade image classifier:
3.2.1 The first one is training and
3.2.2 The other is detection.
OpenCV provides two applications to train cascade classifiers
OpenCV HAAR training and OpenCV train cascade. These two
applications store the classifier in different file formats.
For training, we need a set of samples. There are two types of
samples.
3.2.3 Negative sample: It is related to non-object images.
3.2.4 Positive sample: It is a related image with detects
objects.
A set of negative samples must be prepared manually, whereas
the collection of positive samples is created using the OpenCV
create samples utility.
3.3 LBPH Algorithm
Local Binary Patterns Histogram algorithm is used for Face
Recognition. It is one of the top performing texture descriptors
and is based on the local binary operator. Facial recognition
918
systems are getting more and more important. They are used in
access control, surveillance, and smartphone unlocking, among
other applications. Using LBPH, features from a test image input
are extracted and compared to the database of faces as we see in
Fig 2.
Steps of the algorithm
3.3.1 Selecting the Parameters: The LBPH accepts the four
parameters: Radius, Neighbours, Grid X & Y.
3.3.2 Training the Algorithm.
3.3.3 Using the LBP operation.
3.3.4 Extracting the Histograms from the image.
Fig.2 LBPH Algorithm Working
3.4 Microphone Use for Voice Recognition
They are mentioned implementation uses The Free ST American
English Corpus dataset (SLR45), which is a free American
English corpus by Surfing technology, containing utterances
from 10 speakers (5 females and 5 males)
Once we download the data set, we split it into two different
parts
3.4.1 Training set: Some parts for training the individual
gender models.
3.4.2 Testing set: Some parts for testing the accuracy of
gender recognition.
Voice Features Extraction- The Mel-Frequency Cestrum
Coefficients (MFCC) are used here since they deliver the best
results in speaker verification. MFCCs are commonly derived as
follows
3.4.3 Take the Fourier transform of (a windowed excerpt
of) a signal.
3.4.4 Map the powers of the spectrum obtained above onto
the Mel scale, using triangular overlapping windows.
3.4.5 Take the logs of the powers at each of the Mel
frequencies.
3.4.6 Take the discrete cosine transform of the list of Mel
log powers, as if it were a signal.
3.4.7 The MFCCs are the amplitudes of the resulting
spectrum.
To extract MFCC features I usually use the Python-Speech-
Features library, it is simple to use and well-documented.
Gaussian Mixture Models:
To train Gaussian mixture models based on some collected
features, you can use scikit-learn-library specifically the scikit-
gm as we see the Gaussian Mixture Model in the Fig 3
Fig.3 Gaussian Mixture Model
3.5 Speech to text Phrases Matching
Steps of Working:
3.5.1 Our voice and phrases will be trained through CMU
Sphinx.
3.5.2 With CMU Sphinx voice sample will be transcribed.
3.5.3 With the result that the phrase would be saved in data
for the initial stage.
3.5.4 When the system would be set up completely and user
data would be entered. The input/voice sample
provided by the user would be transcribed and
compared to throughout the data to match the phrase.
CMU Sphinx
3.5.5 State of art speech recognition algorithms for efficient
speech recognition. CMU Sphinx tools are designed
specifically for low-resource platforms.
3.5.6 Support for several languages like US English, UK
English, French, Mandarin, German, Dutch, Russian
and the ability to build a model for others.
3.6 RFID System
The reader/writer device emits electromagnetic waves at a
certain frequency through the antenna when the magnetic card
is in the broadcast area of the reader. It receives the energy and
retransmits its own code. From there, it knows exactly which
devices are in the control area.
Most RFID systems often have multiple read devices
connected to a central computer.
RFID Components:
A simple RFID system is made up of two main components:
device reader/writer and tag as we see in Fig 4.
Device Reader/Writer: It is a wireless communication device
that can detect tags that have the same operating frequency
within a certain range.
919
Fig 4: Raspberry Pi in place of computer.
The RF module here is used along with a pair of
encoder/decoder. The encoder is used for encoding the bits to
be transmitted parallel which are decoded by the decoder on
their reception at the receiver. The encoder/ decoder pair used
in the proposed model is HT12E HT12D. The ‘12’ in the
name means 8-address lines and 4-data lines. The encoder has
four input lines. These lines serve the purpose of providing the
input which has to be encoded. The input given to data pin is in
parallel form which is being transmitted into serial form from
the data output pin. The resistance to be applied between
oscillator pins can be found out from the frequency vs voltage
graphs for decoder and encoder given in Fig 5 and Fig 6,
respectively. [14] As we also add System Control flow chart
which is given in Figure 7.
Fig 5: Frequency of oscillation vs Vcc graph for decoder.
Fig 6: Frequency of oscillation vs Vcc graph for encoder.
Vcc = 9V, frequency of oscillation (encoder) = 3.7 KHz,
frequency of oscillation (decoder) = 185 KHz. From Figure 5
and Figure 6 resistance across oscillator pins (encoder) = 820
KΩ, resistance across oscillator pins (decoder) = 47 KΩ.
Fig 7: Control system is the RFID-based security system.
3.7 PIN Generator
If a user enters a password that is the reverse of the original
password, the silent alarm will go to the police. We are using
the RSA algorithm to encrypt passwords and save them on the
local machine and users can add PINs according to their choice.
Fig 8: Block Diagram of Entrance Door for PIN
In figure 8, it shows the block diagram of the secured
entrance door lock system. In the hardware design and
implementation, five functional units are involved. The
functional units include: the power supply unit, the processing
unit, the display unit, the Electromagnetic Lock (EM) unit and
the matrix keypad unit.
This will be documented, implementation of the capabilities,
strength and Effectiveness of the secured door lock system with
doors used. The expected outcome/result of the research is
920
shown in Table 1. [13]
Table. 1 Test Case for the PIN
4. COMPARATIVE STUDIES
The comparative study presented in this section examines
theoretical issues and simulations carried out with the LBP,
HOG, and both combined algorithms, a database with many
classes and few images per class. Due to certain characteristics
of the approaches, it is crucial to employ both types of
classifiers. [8], [10] The Histogram of Oriented Gradients
(HOG) and the Local Binary Pattern Histogram Algorithm
(LBPH) were used to extract two distinct sets of features from
each facial image (HOG).
The classification recognition rate is performed on the Video
Database with various classifiers, and the results are reported in
Tables 2 and 3. The comparative test is carried out with the two
local facial components using the k-Nearest Neighbours
Algorithm (KNN) and the Support Vector Machine Algorithm
(SVM). LBP is an excellent local characteristic for facial
recognition. However, as anticipated, HOG also outperforms
LBH; for this reason, when we combined the two algorithms, the
output supported positive recognition in the KNN classifier, but
in negative recognition, it demonstrated LBH is superior to other
algorithms, demonstrating our system performs better and
produces good results.[19]
Table 2: Recognition Rate on Video Database with KNN
classifier.
Table 3: Recognition Rate on Video Database with SVM
classifier.
According to the results, we can conclude that LBPH Algorithm
is a more effective technique for facial recognition. Also, Table
4 shows the Accuracy rate of the KNN & SVM classifier, and
the three together regions of experiments show that LBH and
Hybrid both conclude better results.
Table 4: Accuracy Rate of KNN & SVM classifier.
There are just some of those technologies that everyone knows
and has accepted for the system to get better and more accurate
results, but bottom line is that currently, Hybrid is not user
intuitive. While there are a lot of recipes, they are all borderline
useless because they all need to be thoroughly customized to the
point of creating a custom model [11], [12].
5. EXPECTED RESULTS
Write the text from your article in different sections. Write the
text from your article in different sections. Write the text from
your article in different sections. Write the text from your article
in different sections. Write the text from your article in different
sections.
In the expected results, we define how our system actually works
and how accurately it gave the output to the user. We also have
a working system so we also attached some screenshots for more
clarity like how Facial Recognition, Voice Recognition works
and how it responds.
921
Fig 5: Sample results taken from data set testing using LBP algorithm.
Fig 6: Data Gathering
In Fig 5, we shows how the sample results taken from data set
testing using LBPH algorithm.
In Fig 6, here we show clearly how the user data gather and also
how we train the data of the user and store that data in the
database (MySQL). It also shows the System where we add user
details and also train the data of the user.
Fig 7: Train the Recognizer
In the above section which represents Fig 7, shows how system
recognize the train data for feed that face data and respective id’s
of each face to the recognizer so that it can learn.
Fig 8: Final Recognition
In Fig 8 here we also see the actual output of the system how it
recognizes the face of the user using Camera and later on we also
add different components in the system for more security like
RFID, PIN, Voice Recognition.
To ensure the safety of lives and properties, a modern
technology that can overcome the challenges of the traditional
approaches is required. [15] [17] This describes the design and
control strategy of a two-factor authentication security system
based on the technology with efficient control facilities and an
enhanced user interface that can secure the entrance to a house.
6. CONCLUSION
This paper includes a summary review of our proposed model
and how it actually works in real-life and also solves many
problems related to face recognition systems based on HAAR,
and OpenCV. Here we discussed different architectures,
approaches, and methods for training or testing images, and
performance measures of face recognition systems were used in
each study. Every researcher has their own method for
identifying faces in databases or on video in order to address the
issues with prior proposed approaches, but these methods are
still not without their benefits and drawbacks. We now have
RFID, PIN, Facial Recognition, Voice Recognition, and three-
factor authentication (multiple). The method helped to ease a
variety of worries, including the potential for cheating in various
movement and facial expression components. The encryption
method also improves security by preventing unauthorized
tampering with the material that has been recorded.
REFERENCES
1. Hardika, C. and Chitaliya, N. (2017). Smart Attendance Management and
Analysis with Signature Verification. International Journal of Advance
Research, Ideas and Innovations in Technology, 3(3), 89-90.
https://www.ijariit.com/manuscripts/v3i3/V3I3-1150.pdf.
2. Yadav, D. K., Singh, S., Pujari, S. and Mishra, P.(2015). Fingerprint
Based Attendance System Using Microcontroller and Labview.
International Journal of Advanced Research in Electrical, Electronics and
Instrumentation Engineering, 4, 5111-5121, doi:
https://doi.org/10.15662/ijareeie.2015.0406029.
922
3. Al-Naima, F. M. and Saleh, M. A.(2015). A Proposed RFID Based
Student Attendance System. International Journal of Computing and
Network Technology, 3(2), 49-55,
https://link.gale.com/apps/doc/A606881367/AONE?u=anon~d39369d&
sid=googleScholar&xid=3fc2edc7.
4. Walton, C. (1973). Electronic Identification & Recognition System. U.S.
Patent No. 3,752,960: https://patents.google.com/patent/US4388524.
5. Lee, Y. J. et al. (2007). Biometric Key Binding: Fuzzy Vault Based on
Iris Images. Lecture Notes in Computer Science, International
Conference on Biometrics, Berlin, Heidelberg: Springer. 4642, 800-808.
doi: https://doi.org/10.1007/978-3-540-74549-5_84.
6. Kasar, M. M., D. Bhattacharyya, and T. H. Kim (2016). Face Recognition
Using Neural Network: A Review. International Journal of Security and
Its Applications, 10(3), 81-100. doi:
https://dl.acm.org/doi/10.5555/2391659.2391750.
7. Chin, H. (2018). Face Recognition Based Automated Student Attendance
System. Dissertation, UTAR. http://eprints.utar.edu.my/2832/1/EE-2018-
1303261-1.pdf.
8. Nilsson, M., Nordberg, J. and Claesson, I. (2007). Face Detection Using
Local SMQT Features and Split up Snow Classifier. IEEE International
Conference on Acoustics, Speech and Signal Processing - ICASSP '07,
Honolulu, HI, USA, II-589-II-592. doi:
https://doi.org/10.1109/ICASSP.2007.366304.
9. Bifari E. N. and Elrefaei, L. A. (2014). Automated Fingerprint
Identification System Based on Weighted Feature Points Matching
Algorithm. International Conference on Advances in Computing,
Communications and Informatics (ICACCI), Delhi, India. 2212-2217.
doi: https://doi.org/10.1109/ICACCI.2014.6968559.
10. Agrawal S. and Khatri, P. (2015). Facial Expression Detection
Techniques: Based on Viola and Jones Algorithm and Principal
Component Analysis. Fifth International Conference on Advanced
Computing & Communication Technologies, Haryana, India, 108-112.
doi: https://doi.org/10.1109/ACCT.2015.32.
11. Rowley, H. A., Baluja, S. and Kanade, T. (1998). Neural Network-Based
Face Detection. IEEE Transactions on Pattern Analysis and Machine
Intelligence, 20(1), 23-38. doi: https://doi.org/10.1109/34.655647.
12. Reynolds, D. A., Quatieri, T. F. and Dunn, R. B. (2000). Speaker
Verification Using Adapted Gaussian Mixture Models. Digital Signal
Processing, 10(3), 19-41, doi: https://doi.org/10.1006/dspr.1999.0361.
13. Umar, A. O. (2019). A Secured Entrance Door Lock System Using
Password Based. African Scholars Journal of Pure and Applied Science,
15(9),110-126,https://www.africanscholarpublications.com/wp-
content/uploads/2020/06/AJPAS_Vol5_No9-7.pdf.
14. Verma, S., Garg, N., Sharma, S. and Kashyap, I. (2018). A Novel
Approach to RFID Based Automated Parking Charges Collection System.
International Journal of Computer Applications, NCNCCACT 2017, 7-
13,https://research.ijcaonline.org/ncnccact2017/number1/ncnccact20170
08.pdf.
15. Okundamiya M. S. and Emakpor, S.(2017). Design and Control Strategy
of a Security Door System Using Radio Frequency Signal. IEEE 3rd
International Conference on Electro-Technology for National
Development (NIGERCON), 406-412.
https://www.researchgate.net/publication/323228541
AUTHORS
Manjot Kaur Bhatia received the PhD degree
in information security from the University of
Delhi, Delhi, India. She is currently working as
a Professor of Computer Science with the
Jagan Institute of Management Studies, Delhi.
She has more than 20 years of teaching and
research experience in the areas of information
security, databases, Linux, and operating
systems. Her research interests include cloud computing,
steganography, data hiding, information security, and software
testing.
E-mail: manjot.bhatia@jimsindia.org
Gurpreet Kaur completed her
bachelor’s degree in computer
applications from the Jagan Institute
of Management Studies, located in
Delhi in 2021. Currently, she is
pursuing her master’s degree in
computer applications from the same
institution. Driven by her passion for
technology, she has developed a keen interest in various
domains within the field of computer science. Her areas of
expertise and interest lie in Python programming and front-end
development.
E-mail: 156.gurpreet@gmail.com
Rajat Tanwar completed his
bachelor’s degree in computer
applications from the Institute of
Information Technology &
Management in Delhi in 2021.
Currently, he is studying in Master of
Computer Applications at Jagan
Institute of Management Studies, Delhi. He has a strong
passion for technology and finds various aspects of computer
science fascinating. He is skilled in Python programming,
front-end development, and blockchain technology.
E-mail:rajattanwar63@gmail.com
Jayant Marwaha completed his
bachelor’s degree in computer
applications at Chander Prabhu Jain
College of Higher Studies & School of
Law in Delhi in 2021. Currently, he is
pursuing his master’s degree in
computer applications from Jagan
Institute of Management Studies
affiliated to GGSIPU. His profound passion for technology has
ignited a keen interest in various domains within the realm of
computer science, with a particular focus on Python
programming and machine learning.
E-mail: jayant_mca21@jimsindia.org
Mrinal Narang completed his
bachelor’s degree in computer
applications from the Institute of
Information Technology &
Management in Delhi in 2021.
Currently, he is studying in Master of
Computer Applications at Jagan Institute of Management
Studies, Delhi. He has a strong passion for technology and
finds various aspects of computer science fascinating. He is
skilled in Python programming, b Linux and Cloud platform.
E-mail: mrinalnarang51@gmail.com
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
This paper presents a study of current applications of Radio Frequency Identification (RFID) for on-road vehicles. RFID has shown a remarkable performance in the state of art Industrial applications. In recent times, RFID can be considered to be one of the most promising and vital wireless communication technologies. A large gamut of applications employing RFID in moving vehicles is in the nascent stage in different parts of the world. These are location and position of four-wheelers/two-wheelers, Automation of Toll Challan System, Traffic violations, congestion on roads, theft of vehicles, parking allotment etc. An automated e-payment system for parking lots is one of the vital utility that has been presented here and this further resolves in reducing the processing time for the collection of parking payment/charges.
Conference Paper
Full-text available
This paper describes the design and control strategy of a security door system based on the Radio Frequency Identification (RFID) technology. The objective was to provide efficient system monitoring facilities that can secure the entrance to a building. The design utilised a two-factor authentication mechanism as well as a microcontroller programmed in C++ to control the hardware system. An electromagnetic relay was designed to regulate the opening and closing of the door being moved by a synchronous motor. The multiplexers, which consist of integrated circuits, were used for the display of the status of the card user in the liquid crystal display while the alarm system notifies the public of an intruder. The results are presented and discussed.
Conference Paper
Full-text available
Most of fingerprint identification systems perform matching algorithms based on different minutiae details present in the fingerprint. Usually, minutiae are extracted from the thinned fingerprint image. Due to the image noise and different preprocessing methods, the thinned image could results in a large number of false minutiae which may decrease the performance of the system. In this paper, some existing algorithms from different studies were integrated to build Automated Fingerprint Identification System. A new matching algorithm was proposed based on feature with two points; minutiae and ridge point. In addition, each feature extracted assigned to an appropriate weight according to proposed weights table. The modified system was tested on FVC2002 DB1 set-A and all four FVC2004 set- A databases and showed that it is effective and gives excellent results that exceed the performance of classic minutiae-based matching algorithm.
Article
Full-text available
Fingerprint matching has been successfully used by law enforcement for more than a century. The technology is now finding lot of other applications such as identity management and access control. In this context, an automated fingerprint recognition system and identification of key challenges are described along with the research opportunities. The description is like a product design in this report implementing RTOS (Real time operating system) under the domain of embedded system. Fingerprint Recognition is a widely popular but a complex pattern recognition problem. It is very difficult to design accurate algorithms capable of extracting salient features and matching them in a robust way. In this paper, we have come with a novel approach to simplify the existing problems with a proper Embedded System Design KEYWORDS: Microcontroller, Fingerprint, Biometric, Recognition, Embedded system.
Conference Paper
Full-text available
We present a neural network-based upright frontal face detection system. A retinally connected neural network examines small windows of an image and decides whether each window contains a face. The system arbitrates between multiple networks to improve performance over a single network. We present a straightforward procedure for aligning positive face examples for training. To collect negative examples, we use a bootstrap algorithm, which adds false detections into the training set as training progresses. This eliminates the difficult task of manually selecting nonface training examples, which must be chosen to span the entire space of nonface images. Simple heuristics, such as using the fact that faces rarely overlap in images, can further improve the accuracy. Comparisons with several other state-of-the-art face detection systems are presented, showing that our system has comparable performance in terms of detection and false-positive rates.
Conference Paper
Facial expression is a prominent posture beneath the skin of the face. They are the way of communication in humans which convey many things non-verbally. During the past years face recognition has received significant attention as one of the most important applications of image understanding and analysis. Many algorithms have been implemented on different static and non-static conditions. Static conditions include static and uniform background, identical poses, similar illumination, neutral frontal face. Non static conditions include position, partial occlusion orientation, varying lightening conditions and facial hair which make recognition process a complex problem. All these factors influence face recognition process. The main stages for face recognition include face detection, feature representation and classifications. Researchers have described distinct approaches for face recognition. In this work we present a glimpse of face detection techniques, methods used, their performance a their limitations and proposed a new technique for Face Detection based on Viola and Jones algorithm and principal component analysis. At the end we have shown simulation results for the proposed technique and established that proposed technique is performing better than the existing one. The proposed system is implemented in MATLAB version 7.1.4.0.739 (R2012a).
Conference Paper
The purpose of this paper is threefold: firstly, the local successive mean quantization transform features are proposed for illumination and sensor insensitive operation in object recognition. Secondly, a split up sparse network of winnows is presented to speed up the original classifier. Finally, the features and classifier are combined for the task of frontal face detection. Detection results are presented for the MIT+CMU and the BioID databases. With regard to this face detector, the receiver operation characteristics curve for the BioID database yields the best published result. The result for the CMU+MIT database is comparable to state-of-the-art face detectors. A Matlab version of the face detection algorithm can be downloaded from http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId= 13701& objectType=FILE
Article
Reynolds, Douglas A., Quatieri, Thomas F., and Dunn, Robert B., Speaker Verification Using Adapted Gaussian Mixture Models, Digital Signal Processing10(2000), 19–41.In this paper we describe the major elements of MIT Lincoln Laboratory's Gaussian mixture model (GMM)-based speaker verification system used successfully in several NIST Speaker Recognition Evaluations (SREs). The system is built around the likelihood ratio test for verification, using simple but effective GMMs for likelihood functions, a universal background model (UBM) for alternative speaker representation, and a form of Bayesian adaptation to derive speaker models from the UBM. The development and use of a handset detector and score normalization to greatly improve verification performance is also described and discussed. Finally, representative performance benchmarks and system behavior experiments on NIST SRE corpora are presented.
Conference Paper
Recently, crypto-biometric systems have been studied for solving the key management problem of cryptographic systems and protecting templates in biometric systems at the same time. The fuzzy vault system is a well-known crypto-biometric system. We propose a new method of applying iris data to the fuzzy vault. Our research has following two advantages and contributions. First, in order to solve the variation problem of the extracted iris features, we introduce a pattern clustering method. Second, in order to produce unordered sets for fuzzy vault, we use the iris feature extraction algorithm based on ICA (Independent Component Analysis). Experimental results showed that 128-bit cryptographic keys as well as the iris templates were secure with the fuzzy vault scheme.
Smart Attendance Management and Analysis with Signature Verification
  • C Hardika
  • N Chitaliya
Hardika, C. and Chitaliya, N. (2017). Smart Attendance Management and Analysis with Signature Verification. International Journal of Advance Research, Ideas and Innovations in Technology, 3(3), 89-90. https://www.ijariit.com/manuscripts/v3i3/V3I3-1150.pdf.