Content uploaded by Md. Jahidul Islam
Author content
All content in this area was uploaded by Md. Jahidul Islam on Nov 22, 2023
Content may be subject to copyright.
Content uploaded by Md. Jahidul Islam
Author content
All content in this area was uploaded by Md. Jahidul Islam on Oct 18, 2023
Content may be subject to copyright.
147
4SQR-Code: A 4-state QR code generation model for increasing data
storing capacity in the Digital Twin framework
Ababil Islam Udoy
a
, Muhammad Aminur Rahaman
a,
⇑
, Md. Jahidul Islam
a
, Anichur Rahman
b,
⇑
, Zulfiqar Ali
c
,
Ghulam Muhammad
d,
⇑
a
Department of CSE, Green University of Bangladesh, Purbachal American City, Kanchan, Rupganj, Narayanganj-1461, Dhaka, Bangladesh
b
Department of CSE, National Institute of Textile Engineering and Research (NITER), Constituent Institute of the University of Dhaka, Savar, Dhaka 1350, Bangladesh
c
School of Computer Science and Electronic Engineering, University of Essex, United Kingdom
d
Department of Computer Engineering, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi Arabia
highlights
A novel 4SQR code is proposed based
on traditional QR codes.
We implement the 4SQR to generate
and detect the readability of the
system.
The proposed 4SQR can increase data
storing capacity with security.
It is tested in a digital twin
framework using randomly generated
samples.
graphical abstract
article info
Article history:
Received 27 May 2023
Revised 13 September 2023
Accepted 6 October 2023
Available online xxxx
Keywords:
Quick response (QR) code
4-State QR (4SQR) Code
Digital Ttwin
Encoding
Decoding
abstract
Introduction:The usage of Quick Response (QR) Codes has become widely popular in recent years, pri-
marily for immense electronic transactions and industry uses. The structural flexibility of QR Code archi-
tecture opens many more possibilities for researchers in the domain of the Industrial Internet of Things
(IIoT). However, the limited storage capacity of the traditional QR Codes still fails to stretch the data
capacity limits. The researchers of this domain have already introduced different kinds of techniques,
including data hiding, multiplexing, data compression, color QR Codes, and so on. However, the research
on increasing the data storage capacity of the QR Codes is very limited and still operational.
Objectives:The main objective of this work is to increase the data storage capacity of QR Codes in the IIoT
domain.
Methods:In the first part, we have introduced a 4-State-Pattern-based encoding technique to generate
the proposed 4-State QR (4SQR) Code where actual data are encoded into a 4SQR Code image which
increases the data storage capacity more than the traditional 2-State QR Code. The proposed 4SQR
Code consists of four types of patterns, including Black Square Box (BSB), White Square Box (WSB),
Triangle, and Circle, whereas the traditional 2-State QR Codes consist of BSB and WSB. In the second part,
the 4SQR Code decoding module has been introduced using the adaptive YOLO V5 algorithm where the
proposed 4SQR Code image is decoded into the actual data.
https://doi.org/10.1016/j.jare.2023.10.006
2090-1232/Ó2023 The Authors. Published by Elsevier B.V. on behalf of Cairo University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
q
Peer review under responsibility of Cairo University.
⇑
Corresponding authors.
E-mail addresses: aminur@cse.green.edu.bd (M.A. Rahaman), arahman@niter.edu.bd (A. Rahman), ghulam@ksu.edu.sa (G. Muhammad).
Journal of Advanced Research xxx (xxxx) xxx
Contents lists available at ScienceDirect
Journal of Advanced Research
journal homepage: www.elsevier.com/locate/jare
Please cite this article as: A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al., 4SQR-Code: A 4-state QR code generation model for increasing data storing
capacity in the Digital Twin framework, Journal of Advanced Research, https://doi.org/10.1016/j.jare.2023.10.006
Results:The proposed model is tested in a Digital Twin (DT) framework using randomly generated 3000
testing samples for the encoding module that converts into 4SQR Code images successfully and similarly
for the decoding module that decodes the 4SQR Code images into the actual data.
Conclusion:Experimental results show that this proposed technique offers increased data storage capac-
ity two times than traditional 2-State QR Codes.
Ó2023 The Authors. Published by Elsevier B.V. on behalf of Cairo University. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction
The Industrial Internet of Things (IIoT) and other industry sectors
are increasingly employing QR Codes in this fourth industrial revo-
lution due to their easy use and wide range of application areas
[1]. Humanity needs a better way to use the existing technology
since we are mostly reliant on it, thus researchers are seeking a
straightforward method of information sharing or storage. During
this procedure, we store our data in a quad image, also referred to
as a QR Code. Nearly 6.648 billion people, or 83.04% of the world’s
population (those who possess a smartphone), can utilize a QR Code,
a sort of matrix bar Code that is used to store data quickly. This code
can be useful in many different fields in modern days such as bank-
ing systems, e-commerce, transferring information through integra-
tion in images, industrial facilities, shopping malls, shipyards, food
shops, billboards, magazines, and electronics product [2].
On the demand of the era, another recent technique is the dig-
ital twin (DT), which emerges with the vast possibility of solving a
complex problem. A DT is a virtual replica of a physical object or
system that can be used to test different issues with ease [3]. The
advance of modern technologies like cloud computing, Artificial
Intelligence (AI), Big data, blockchain, and IoT integration with
DT has allowed us to create a physical twin of current status
[4,5]. The research aims to use DT technology to optimize the QR
Code system and bring new 4SQR. The DT can be used to improve
the accuracy and efficiency of the 4SQR Codes, such as improving
generation time, reducing the decoding time, and enhancing the
accuracy, efficiency, and reliability of the 4SQR. However, Fig. 1
depicted excellent collaboration scenarios between a real QR Code
system with the DT techniques that are currently used in the mod-
ern world [6].
On the other hand, traditional QR Fig. 2(a) depicts the structure
of the QR Code [7]. There are several aspects of a QR Code. These
include the code design, the fixed pattern, format information,
encode mode, length of the data (Len), timing pattern, and finally,
the data pattern. Furthermore, a QR Code can only store a limited
number of bits in the image of the existing version of the QR Code.
We come up with a way to improve the data storage capacity of the
QR Code by double(2x) the current standard QR Code by introduc-
ing a new technique to the industries by proposing a new system.
We get to a new horizon of data storing in QR Codes. In the field
of QR Codes, there has been a lot of work done by researchers in the
last two decades [8], but there are lots of things that are still lack-
ing. such as limited data storage, limited error-correcting system,
limited usability, etc., The Two-State Pattern Technique (BSB &
WSB) is adopted in the industry where we have introduced the
"4-State Pattern Technique”. The existing two-state has a masking
part, but if the masking part gets distorted, it makes the whole QR
Code unreadable. This is why we introduce the 4-State Pattern
Technique QR Code which removes the masking pattern. Industrial
costs can be minimized by the proposed technique. Moreover, QR
Codes have some essential uses in the industry, but existing QR
Codes have less data storage capacity which we can increase
through our proposed "4-State Pattern Technique based Method”.
Therefore, we aim to create a QR Code for collecting information
with minimum time and computational cost and, increase the
accuracy of the existing system. In the proposed system, several
test cases are generated to test the data storage capacity, time
complexity, and memory complexity. Our main aim is to reduce
the limitations of convolution when increasing the data storage
capacity and prepare a simple intellectual way to ensure all the
techniques that have been followed for increasing capacity.
Research questions & contributions of the study
Q1. How can we increase the storage capacity of the proposed
4SQR-Code system?
Q2. How can we easily use an 4SQR-Code system and what are
the test cases?
Q3. How much accuracy does an 4SQR-Code system provide?
Q4. How to improve accuracy and minimize computational cost
and what are those different approaches?
The contributions of this paper are as follows:
We have designed a structure for increasing existing system
data storing capacity.
We have implemented the 4SQR-Code to generate and detect
the readability of the system.
Also, a 4SQR-Code algorithm is proposed for increasing data
storing capacity.
In addition, we evaluate the data storing capacity and perfor-
mance of the proposed model.
Organization
The rest of the paper has been formed as follows: We have stud-
ied and discussed the literature review in Related Works section.
After that, Proposed 4-State QR Code Pattern section presents the
structure for 4SQR; architecture for 4SQR; procedure for 4SQR.
Moreover, an evaluation of the results and analysis of the proposed
model are provided in Experimental Results And Discussion sec-
tion. In sum, the authors concluded the paper in Conclusion section.
Related works
In this section, we have gone through some recent research
works such as data hiding techniques [9–11], multiplexing tech-
niques [12–14], use of color code techniques [15,16], use of data
compression techniques [17,18], etc., but most of them are focused
only on the outer level of the QR Code system instead of a more
root level approach. QR Codes are used in several types of domain
applications, particularly when transferring various types of data
between electronic devices is needed. In essence, the fourth indus-
trial revolution is the trend towards automation and data exchange
[19] in manufacturing technologies and processes which include
Internet of Healthcare Things [20], IoT [21], industrial Internet of
things [22], cloud computing [23–25], cognitive computing, and
artificial intelligence. Following automation and data exchange
technologies, the QR Code might be a fantastic communication
channel between many electronic gadgets. It means that the QR
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
2
Code requires a large amount of data storage capacity in order to
exchange a large amount of data, which is possible if we increase
the QR Code’s data storage capacity. The rising data storage capac-
ity may aid in meeting the needs of current technology. The
author’s objective in our suggested system is to expand the data
storage capacity of the present QR Code technology. As a result,
we should familiarize ourselves with the conventional QR Code
system. There is also a need to learn about various strategies used
in this sector, such as data concealing techniques, multiplexing
techniques, color coding techniques, data compression techniques,
Fig. 1. Scenario of DT technology with real QR Code System.
Fig. 2. (a) Traditional QR Code and (b) Proposed 4-State QR Code Structure.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
3
and so on. We analyzed their approach in order to improve the
work progression of concept implementation in our system.
Most systems have primarily concentrated on various methods
(compression techniques, color code techniques, zip encoding) to
tackle the limited storage issue, and they are not nearly there; as
a consequence, we cannot consider their methodology to be the
QR Code’s successor.
Digital twin
The concept of DT originates from Michael Grieves in the year
2002 at a conference [3]. As the conceptual model guiding product
life-cycle management, Grieves offered the DT. The actual product,
its virtual representation, and a bidirectional data link between
them make up DT, in accordance with his notion. In the context
of product life cycles, such as simulation, integration, testing, mon-
itoring, and maintenance, this concept focuses on DT. The DT
method offers a crucial basis for linking physical items to their vir-
tual counterparts, enabling greater learning and interaction over
time and space. By incorporating third-party services, it does more
than only assist in improving design and operations; it also estab-
lishes a new industrial ecosystem. For businesses like manufactur-
ing, the change from a purely product-focused perspective to a
more industrial one is crucial [26–31].
Compression techniques
We have analyzed a few compression techniques that have
been used for the QR Code system, which has been focused on
increasing the capacity of QR Code by applying different compres-
sion techniques [32,17,33]. After investigating numerous tech-
niques, we have identified that there are some different
parameters that are mainly focused on various models and algo-
rithms like Huffman encoding [34], zip encoding, etc.; these types
of techniques are the most common in the area which makes a
great impact on visualizing the importance and areas of increasing
capacity of QR Code. Ali et al. [17] have developed a system that
enhances the QR Code capacity using a lossless compression tech-
nique, and they recognized the verification of secure e-documents.
Their system mainly focused on using different hash values for
integrity ensuring and compression using Huffman encoding,
which is a lossless data compression algorithm. The main focus
of this system was enhancing the QR Code capacity, which was sat-
isfied by the feature of satisfying the security requirements. More-
over, using compression technique for QR Codes focusing the
increasing capacity has been developed by Arora et al. [18], which
system mainly focused on using zip encoding for data compression
algorithm and then color QR Code for applying modified multi-
color QR encoding technique. The main focus of this system was
increasing the storage capacity, which resulted in a maximum of
29% increment with a difference of 14% to the minimum based
on different levels of QR Code. We have noticed that the increasing
storage capacity is very much convoluted between BSB & WSB QR
Code and multi-colour QR Code. We have identified different
parameters by analyzing some different compression techniques
focusing on increasing capacity [35]. We have distinguished the
convolution of storage capacity when using color instead of black
and white. We also spot that there are some great techniques that
enhance the capacity of QR Codes, which is very important. For
these reasons, we have to find a better way to consider those
two in a way that the techniques can fulfill their requirement
and also remove the convoluted limitations.
Encoding–decoding techniques
There are various types of work that have been done focusing
on encoding–decoding techniques. We have analyzed a few encod-
ing–decoding related papers [14,36,17]. After investigating several
methodologies, we discovered multiple parameters associated
with various models and algorithms, such as Huffman encoding,
multiplexing, and lossless compression techniques. Umaria et al.
[32] Enhancing the data storage Capacity in QR code using Com-
pression Algorithm and achieving security and Further data storage
capacity improvement using Multiplexing. This paper proposes a
technique for increasing data capacity by zip compressing and
multiplexing and retrieving data by reversing. By using zip com-
pressing and multiplexing, the system creates a QR Code with
increased data capacity and provides data security. Another system
developed by Sijia Liu et al. [37] Rich QR codes with three-layer
information using Hamming code. This paper introduces a (n,n)
secret sharing scheme, where n¼2
p
. In this setup, there are three
main roles: a secret distributor, a secret compositor, and n partic-
ipants. The secret distributor encodes the second and third-layer
information into multiple QR code shares, all of which can be accu-
rately decoded using a standard QR code reader. During secret
recovery, the second layer information is obtained through XOR
operation, followed by the extraction of the third layer
information.
Color code techniques
There are various types of work that have been done focusing
on using the color QR Code technique. We have analyzed a few
color QR Code that has been used for the different type of QR Code
[38–42]. After investigating those systems, we have identified that
there are different models and approaches for different QR Codes.
Those different systems have various working processes. The sys-
tem developed by Taveerad et al. [38] for the development of color
QR Codes for increasing capacity. Moreover, they try to convert the
bit stream into code ward hexadecimal [43] to use a different color
than that hexadecimal code converts into binary. Those binary val-
ues are placed into the QR matrix and generate images. Galiyawala
et al. [39] to increase the data capacity of QR Code using multiplex-
ing with color-coding: An example of embedding speech signals in
QR Code. This paper proposes a technique for increasing data
capacity by multiplexing more QR Codes in the same version using
color coding. Moreover, the proposed method has several black-
and-white QR Codes that are multiplexed together. For every dis-
tinct binary pattern, a distinct combination of RGB (Black, Blue,
Green, Red, Cyan, Magenta, Yellow, and White) weights is assigned
to its new QR Code. This will generate a multiplexed single-color
QR Code with increased capacity. By analyzing different color QR
systems [40–42,44], we have identified that there are very few
common patterns. Most of the patterns are different from one
another and have distinct processes for encoding and decoding.
For those reasons, we select a root-level pattern technique to
ensure efficiency.
Standard technique
In the field of QR Code [7] ISO/IEC 18004:2015, ”Information
technology–Automatic identification and data capture techniques,
2015(E)”. There are not many common patterns, but a few com-
mon patterns exist. Those common patterns used by most QR
Codes, those common patterns are converting characters to binary,
Solomon encoding, and converting a matrix to an image. All of the
systems above use those common patterns.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
4
Comparison with existing work
There are different types of techniques that are used by various
types of systems like zip encoding, multiplexing, data embedding,
Huffman coding, Solomon encoding, etc. For the need to increase
the capacity of QR Codes. We identified that there are different
type of convolution that happens while increasing data storage
capacity. We also observe that there is a decrease in performance
between the BSB and WSB or standard QR Code and Multi-Color
QR Code [38,45] which indeed resulted in a focusing point for
the proposed technique. There are very few common patterns
where most of the patterns are different from one another but
Fig. 3. The system architecture of the proposed 4-State QR Code pattern generation (encoding) and recognition (decoding) based on DT which is presented by Fig. 4 in detail.
Fig. 4. DT Model for the proposed 4-State QR Code pattern.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
5
many techniques of color QR Code increase the importance of these
techniques [38–40,44] which indeed is resulted as a selecting key
to going on a root level pattern technique for the proposed
technique.
However, different literature reviews with a focus on increasing
the capacity of QR Codes with techniques like compression and
color code are mainly discussed. Those techniques have used vari-
ous parameters and algorithms. Their proposed algorithms were
Fig. 5. 4-State QR Code image generation from NxN matrix where four patterns are represented by 00, 01, 10, and 11.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
6
discussed in the schemes of several aspects, such as secrecy[46–
52], durability, complexity, and storage capacity [7]. They have dif-
ferent strategies used to expand the capacity limit according to the
business request. All of the papers shows various approach to
improving the capacity of QR Code, but we can see carefully that
most of the system process is nearly similar in type of working pro-
cedure but different in the process. When there are numerous
types of processes going on in various models, we can see the dif-
ference between the complexity of the data and the capacity of
data storage. On the other hand, that system’s better performance,
usability, and efficiency depend on the different system’s working
processes. When we thought about working with them and giving
them an extra opportunity so that they could have some more
areas to work with, we thought about the proposed system.
Proposed 4-state QR code pattern
System optimization is a process where the system gets opti-
mized. For the process of optimization, creating a new system is
very important. If we propose a new technique for QR Codes, then
we also need to introduce a special method of testing. This paper
proposed the ”4-State Pattern Technique” to ensure the optimiza-
tion of data storing capacity and getting output as decoded data.
The proposed method works for the QR Code generation system
and also for the QR Code decoding system. In this system, there are
2 (two) main parameters, one is encoding, and another is decoding.
where the encoding module creates a ”4-State QR Code” and the
decode module decodes the QR Code image. We implement the
procedure from standard ”ISO/IEC 18004:2015 of QR Code” [7].
Fig. 2(b) shows the structure of the proposed system, which fol-
lows the proposed four-state pattern technique. In our proposed
system, there are mainly two parts which are: encoding and decod-
ing where there are several processes that need to be followed
according to the architecture. Fig. 3 in the encoding section main
idea is to use two bits by two bits-by-bits binary data-splitter to
split the data. Then we apply it to the matrix value placer and by
doing so we get different patterns with respect to given data.
Matrix to image generator places every two bits to blocks where,
WSB (as binary number = 00), BSB (as binary number = 01),
Fig. 6. Image Patch comparing and value assigning to Matrix in 4-State Pattern Technique.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
7
Triangle (as binary number = 10), and Circle (as binary number
= 11). Now we will get an output of the QR Code. (see Fig. 4).
Fig. 3, in the decoding section, the main idea is to decode actual
data from a QR Code image. For this, the image is first sent to an
image patcher after scanning it creates a data matrix from the
image patcher data. Then sent to image-data-extraction data sepa-
ration. after, data goes to the decoder to decode the Solomon[53]
encoded data. Now the data gets separated and converted from
digit to utf-8 character. Finally, we get an output which is the
actual data. Moreover, Fig. 3 shows the architecture of the pro-
posed system which follows the proposed "4-state pattern tech-
nique”. where there are several processes that need to be
followed accordingly with the diagrams.
In Fig. 3 in the encoding section, first the input module takes
user data as input and passes that data to two different module
data length modules and the iso-8859–1 encoding conversion
module. The data length module finds out the length (how long
thedatais?)ofthedata,andtheiso-8859–1encodingconversion
module converts data to digits. After doing the iso-8859–1
encoding conversion module it passes its data to two(2) modules,
one is Solomon encoding and another is Conversion to binary.
Solomon encoding converts its received data into polynomial
data and generates digits. After that data length, iso-8859–1
encoded data, and Solomon encoded data converted into binary
data. The system passes those data to 2x2 bits binary data splitter
and creates a data group such as ”10 11 00 11 00 01 10 11”. Par-
allelly data length passes its data to the matrix size finder, which
finds the accurate matrix for the given data and passes data to a
module called matrix generator. The matrix generator builds a
matrix filled with all the necessary fixed modules. For the final
matrix generation, a module named matrix value placer gets
2x2 bits binary data splitter value and places those values into
the matrix. Finally, Fig. 5 shows that there is a module called
matrix to image generator, It creates an image by taking value
from the matrix and pre-defined shapes/pattern of the images
and creates a final image.
Fig. 3 shows the diagram of decoding, the first module is the
input-image module that takes an image as input. Then passes into
the image patcher, which decides the patch size and creates mini
patches of the image. Then we applied our custom-designed adap-
tive YOLO v5 Algorithm, From that system gets the accurate size of
an individual patch of images.
Fig. 6 shows the similarity measurement procedure of how
those patches of the image go through a custom image compara-
tor and compare those image with a system-predefined image to
see their similarities after finding the highest number of similar-
ities in which image get the highest number of, the predefined
image has its corresponding integer value. Those integer values
convert into 2(two) 2-digit binary values and are placed in a
blank matrix. Finally, the system gets a matrix filled with binary
values. After that image-data extractor goes through a specific
way to get the data from the matrix and store it in a data stream.
The next module is the data separator, which separates data into
2(two) blocks, one is the ”data block” and another is a ”Solomon-
encoded data block”. After that next module decodes Solomon-
encoded data and returns a data stream. From there data block
& decoded data get compared, after comparing those output data
and Solomon-decoded data. If compare result is similar, then go
to the next module, else correct those error bits. finally, the final
module is the digit to utf-8 character conversion module, in this
state module system converts the data into its corresponding utf-
8 character and returns it as output.
Algorithm 1. Algorithm of 4SQR-Code encoding
Proposed Algorithm 1’s Procedure
Step 1: First take input that is going to use to encode.
Step 2: Find the Data length.
Step 3: Convert the character to a digital character.
Step 4: Apply Solomon Encoding.
Step 5: Converts steps 2,3,4 values into binary values and adds
them to a list.
Step 6: Split those binary values 2 bits by 2 bits.
Step 7: From value finds the matrix size & generates a blank
matrix.
Step 8: Apply all the necessary patterns (version info, format
info, position pattern, alignment pattern & timing pattern) to
the matrix.
Step 9: Apply list value to matrix & convert matrix into image.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
8
Algorithm 2. Algorithm of 4SQR-Code decoding Where, N
i
= Number of string, F
i
= 4SQR image, i= Counter,
L= Length, RS
enc
= RS encoded data, KL
i
=keylist½,D
d
=
digitaldata½,D
i
=Data½,Sd
i
= Splitted data, M
s
= Matrix size, Ma s
= Matrix_size(), M= Matrix, M
g
ðÞ = Matrix_generator(), F
p
ðÞ =
Fixed_pattern(),
a
i
= Apply data, d
i
= image_generation(), I= Image,
D
S
= data stream, PI = Predefined Images[I0,I1,I10,I11], Img = image,
Im
a
= image array, M
i
= Main image, b
i
=remo
v
eborderðÞ,cmp
r
=
compare result, fbs = formatted binary stream, RS
d
= RS data, bts
=binary to string,stcðÞ =strcon
v
ertðÞ and d
d
= decoded data.
Proposed Algorithm 2’s Procedure
Step 1: First take an image as input.
Step 2: Find a single shape of the image. and create an n*n
image patch.
Step 3: Match image patches with predefined images and place
values into the matrix.
Step 4: Check matrix size. If valid then go to the next step, if not
valid repeat step 2 & 3 to find a valid matrix.
Step 5: Extract data from the matrix and split data into 2 parts,
the data part, and Solomon’s encoding(rs) part.
Step 6: From the data part, extract the encoding mode and data
length
Step 7: Decode the Solomon encoding using the rs decoding
technique. Convert values to characters using UTF-8.
Step 8: Match rs code with data, if match successfully then dis-
play data else correct wrong data with Reed–Solomon error cor-
rection technique if Reed–Solomon error correction technique
fails repeat from step 1.
Experimental results and discussion
In the experiment part, we have used 12 GB RAM and an Intel
Core i5 with 7
th
gen processor with a base speed oimageGHz. We
have also used pycharm and spyder for Python and OpenCV wrap-
per, imageIo wrapper, and image_similarity_measures wrapper
have been used. Moreover, 64-bit Operating Systems with Win-
dows 10
Ò
and Ubuntu 20.04 LTS have been used as an implemen-
tation platform. For testing purposes, we have created a few test
cases and tried those test case scenarios to see whether our system
can perform accurately or not. We have generated a list of 4SQR
Code samples to test our system. Furthermore, we have shown a
competitive analysis of memory and time in Table 1.
Performance metrics
Performance evaluation is the assessment of computer systems
which includes their components, processes, and outputs. In per-
formance evaluation there are a few parameters, those parameters
are response time, latency, speed, throughput, and memory usage.
Response time: response time is the time taken by a system to
do a task and return the result.
D
t¼Mþkp
a
ð1Þ
start time ¼
D
t
0
end time ¼
D
t
1
time difference
D
t
D
¼
D
t
1
D
t
0
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
9
Where,
D
t= Time of a position, M = Mean anomaly, k= The ecliptic
longitude of the periapsis, a= The right ascension of the apparent
sun, Start time =
D
t
0
and End time =
D
t
1
.
Speed: The term speed is usually referred to the clock speed of
the processor. It means how fast a processor can complete its task.
We have tested our system on different machines and also found
different results in different machines Because machine-to-
machine speed defers.
Memory usages: Memory is the module where load and run
applications(memory means primary memory). How much mem-
ory is needed in a system to run an application of that system?
That’s why memory is an essential parameter. In this system, we
have taken that into consideration. We have also analyzed our sys-
tem memory usage in this section.
M
U
¼M
T
M
F
ð2Þ
M
P
¼MAXðM
U
Þ
)Peak memory usages ¼M
P
Where, M
U
= Total used memory, M
T
= Total Memory, M
F
= Free
Memory, MAXðM
U
Þ= Maximum memory usage from a group of
sample.
Testing Data
In the system architecture of the proposed 4SQR, we have
observe two(2) input, so for this proposed 4SQR technique author
generate some testing data.
In the fields of QR Code, there are not many test cases, and
mainly those test cases are in the input data stream and input
images. On those, there are not many places for bugs/errors to hap-
Table 1
Test data & images of our encoding proposed system.
Input Data Asif Mahmud Rial Rokibul Islam Ababil Islam Udoy
Length 16 13 17
Maximum Processing Memory
Consumption (MB)
115.76 115.76 115.76
Execution Time (sec) 1 1 2
Output
Input Data Dr. Muhammad Aminur Rahman Green University Of Bangladesh I love my country
Length 26 30 17
Maximum Processing Memory
Consumption (MB)
115.76 115.76 115.76
Execution Time (sec.) 1 1 1
Output
Input Data b3701e7c184b0bbf71
19752603314c49a690cb26 e5f37c1b03f961a4b67ed44
d6e9534d9b73665f9 be66ec47c0176381bc
a1cbc348c5111e6aef1656
Length 40 40 40
Maximum Processing Memory
Consumption (MB)
155.22 155.22 155.22
Execution Time (in sec.) 1 1 1
Output
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
10
pen, those test cases are analyzed in this section. If we follow those
test cases and create test data, then there is no place for errors to
occur in our system. In the performance evaluation parameters,
we have shown some of our test data, on those test data we have
used only ”English Characters” because we have developed our
system only for ”English Characters”. We have processed our data
by its length and sorted those data. After we have collected our
system run-time and main memory cost. Moreover, we have dis-
cussed in result in the result of the proposed system section. From
those test results, run time, and memory cost we have come to a
position where we could say our proposed system works
adequately.
In this work, image generation is a testing case Fig. 3 as shown
in our proposed system. There is the use of image generation as a
testing case to test our system. When any data gets input in the
proposed system Algorithm 1 then the data is converted into an
image. We generated 3,000 testing samples, and then we ran the
encoding system for 3,000 different data sets and got the required
sample. These samples are used for the next test case to see how
the system hold up to different test case. The generation of an
image is the process of image generation in Table 1 as shows that
the image goes to a module and different sub-modules proceed
with their individual task received from the system module. An
image generation system should be robust to any changes in data.
Because when taking sample data, data always may not be in the
clear string, the system needs a clean data stream as input.
In the testing section, scanning is also a test case, as expressed
in Fig. 3. Our system reads the file and creates mini patches with
custom YOLO v5, we have tested our patches with the predefined
image. Moreover, our system works with those patch system to
create a matrix, and from that matrix system get output. We have
tested our system externally and graphed it with standard QR sys-
tem output. The scanning of the image is the process of image
reading, in the image (YOLOv5), goes and gets read and converted
the image to the matrix. Next, different sub-modules proceed with
their individual task like data extraction, data separation, data
decoding, and converting data stream to UTF-8 characters and
complete all the processes received from the system module. An
image reading system should be robust to any changes in the
image. When taking sample image data, image data may always
Table 2
Test image & output data of our decoding proposed system.
Input Image
Execution Time (in sec.) 5 5 5
Maximum Processing Memory
Consumption (MB)
9.597498 9.597498 9.597498
Length 16 13 17
Output Asif Mahmud Rial Rokibul Islam Ababil Islam Udoy
Input Image
Execution Time (in sec.) 5 5 4
Maximum Processing Memory
Consumption (MB)
9.469864 9.469864 9.469548
Length 26 30 17
Output Dr. Muhammad Aminur Rahman Green University Of Bangladesh I love my country
Input Image
Execution Time (in sec.) 7 7 8
Maximum Processing Memory
Consumption (MB)
13.182203 13.183273 13.182011
Length 40 40 40
Output b3701e7c184b0bbf711975 e5f37c1b03f961a4b67ed4 be66ec47c0176381bca1c
2603314c49a690cb26 4d6e9534d9b73665f9 bc348c5111e6aef1656
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
11
come in size and condition, and the system needs to process it as
input. We can observe in Table 1. The raw string is used to generate
the "4-State QR Code” and those QR Code images used in decoding
Table 2.
Table 3 depicts the result of the proposed system and existing
system, The existing system’s data capacity is shown and on the
other hand, we have shown the result of our proposed system.
According to their study, they found a complex growth of their
data, and we have also found complex growth of their data we
have compared the existing system and our system data. We have
designed our system in the base of Byte Mode, where we have dou-
ble Byte mode data in our system.
Character capacity comparison
Fig. 7 depicts the character capacity of the QR systems (Conven-
tional QR vs. 4SQR) compared to the number of various versions (5
to 40) of the QR Code system. At the beginning of the graph, there
are no differences between the existing and proposed systems.
Also, for a small number of versions, proposed and existing sys-
tems are almost similar. However, with increasing the number of
QR versions (Over 10 versions), character capacity also increased
linearly. However, it is clear that the proposed 4SQR Code system
has better character capacity compared to the conventional QR
Code system. For the 40 versions of the QR Code, the proposed
Table 3
Comparative analysis of character capacity(Existing vs. proposed system
Character Capacities
Version Error Correction Level Numeric Mode Alpha numeric Mode Byte Mode Kanji Mode Proposed System
1 L 41 25 17 10 34
M342014828
Q271611722
H17107414
2 L 77 47 32 20 64
M6338261652
Q4829201240
H342014828
.. . . ...
.. . . ...
.. . . ...
39 L 6743 4087 2809 1729 5618
M 5313 3220 2213 1362 4426
Q 3791 2298 1579 972 3158
H 2927 1774 1219 750 2438
40 L 7089 4296 2953 1817 5906
M 5596 3391 2331 1435 4662
Q 3993 2420 1663 1024 3326
H 3057 1852 1273 784 2546
Fig. 7. Character Capacity Comparison with respect to the Versions.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
12
4SQR-Code system has reached a peak of 6000 character capacity
whereas conventional QR is on 3000 character capacity which is
almost half of the proposed system. Overall, it can be seen that
the character capacity of the proposed system is much better than
the conventional QR system.
In Table 4, we have shown what number of versions and what
number of characters conventional QR Codes can create and what
QR can’t be created. On the other hand, we test without data that
whatever it is possible to create double the data on a specific ver-
sion and we show that result in the Table 4. Finally Table 4 depicts
that our system can perform its task as it intended. Also, Fig. 8
depicts traditional QR codes that fail to store large amounts of data
but the proposed system can store the large data.
Fig. 9(a) depicts the complex timing growth of the QR systems
(Standard QR vs. 4SQR) compared to the number of nodes. For
the less (10 to 50) numbers of data, QR and 4SQR take 60 ms and
160 ms respectively. With the increasing of No. data both QR sys-
tems time(ms) also increasing. Moreover, we have observed that
on average conventional QR takes around 65 ms on the other hand,
the proposed system takes around 200 ms. It’s a little bit behind
the conventional QR Code system but this is not much behind as
a newly proposed system, the reasoning is that the proposed sys-
tem works instantaneously for the user. Also, the reason for taking
more time is that the proposed system needs to convert the gener-
ated binary into a 2x2 bit split binary, and for those 2 digits binary,
proposed systems need to create a unique pattern; generating
those pattern take time and memory. (see Fig. 10,11).
Fig. 9(b) shows the memory cost(MB) of the QR systems (Con-
ventional QR vs. proposed 4SQR) compared to the number of
nodes. From the beginning of the graph, there is a big difference
in memory cost. At first, we noticed that standard QR takes around
1 MB to 2 MB of memory, on the other hand, the proposed system
takes around 98 MB to 130 MB of memory. For 1000 to 3000 num-
bers of data, this graph shows a straight line for the proposed QR
Code system, on the contrary, the standard QR system shows a
straight line from the very beginning. The reasoning is that conven-
tional QR doesn’t need to work with the extra module to create a
QR Code, on the other hand, the proposed system needs to follow
a few extra procedures to create a 4SQR Code with much more
data. For that reason, In Fig. 9(b) we have observed a high con-
sumption of memory in the proposed system.
Fig. 9(c) depicts the decoding time of the QR systems (standard
QR vs. proposed 4SQR) compared to the number of nodes. From the
graph, we have observed that conventional QR takes around 50 ms
to 70 ms time(ms), The reasoning is that conventional QR doesn’t
need to go through many processes such as image patcher and
image comparator. Conventional QR first converts the image into
a grayscale/binary raster image and collects data for that reason
to take less time. On the other hand, the proposed system has to
go through an image processing procedure and use custom YOLO
v5 to create a perfect patch, and from those patches, the system
compares the patches’ values and does other possessing and pro-
duces data, for that reason, 4SQR takes much more time compared
to conventional QR system. However, The 4SQR is much more
accurate in contrast to the conventional QR due to image process-
ing that takes much more time to operate.
Fig. 9(d) depicts the decoding memory of the QR systems (Con-
ventional QR vs. proposed 4SQR) compared to the number of vari-
ous nodes. For less no of (50 to 100) nodes, the standard QR Code
takes less than 1(MB) of memory, on the other hand, the proposed
QR Code takes 13(MB). Moreover, after 500 no of nodes, the 4SQR
Code is increased linearly, on the contrary, there is little change
observed in the standard QR Code system. However, in this graph,
we have also observed that conventional QR beat our system by far,
but it’s not a problem for current devices, the reason is that nowa-
days current devices use much more memory than the proposed
system requirement. For that reason, it is not a matter of worry
but in future studies, we can optimize the proposed system so that
the 4SQR system can perform better with much less memory.
Applications
During the COVID-19 pandemic [54], many restaurants began
the use of QR Codes to provide a paperless menu to ensure con-
tactless and safe dining.
It can be applied for product packaging, process management,
product stocking & product picking such as luggage Tags.
Table 4
On which no of data and specific version proposed system can generate 4SQR and QR
can’t
NO of Data OR Character Capacity Version QR 4SQR
17 1 UU
25 1 X U
34 1 X U
32 2 UU
40 2 X U
62 2 X U
Fig. 8. Example outputs of the comparison of data storage capacity (a) Traditional QR codes fail to store large amounts of data (b) The proposed system (4SQR Code) can store
the large data.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
13
Fig. 9. (a) Encoding Time Cost, (b) Encoding Memory Cost, (c) Decoding Time Cost, and (d) Decoding Memory Cost(MB) of the proposed System with respect to standard QR
Code.
Fig. 10. Example output of the proposed system for Encoding process.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
14
In many places people use bar Codes to scan products and also
people use their hands to generate invoices.
Nowadays, QR is used for Point of Sale (POS) such as American
Express, MasterCard, Bkash, Rocket, and Nagad which intro-
duced a feature on their mobile banking apps that enables sell-
ers to accept payments through QR Codes.
Moreover, Social Media such as Facebook, WhatsApp, and Snap-
chat allow users to follow accounts by scanning QR Codes.
Furthermore, in the medical field application of QR is getting
border, and entry of the contents of prescriptions is gotten
smoother through the reading of QR Codes printed on
prescriptions.
Application scenes in the leisure field like walking in museums
and historical sites can give users field information just by scan-
ning it.
Creates ads for mobile applications through 4SQR-Codes. For
instance, Chinese Organizations/ corporations launch their
application through drone shows, they show QR to scan and
download their application to the audience.
We can use the 4SQR Code in libraries to track our books.
Limitations and open challenges
4SQR has introduced only two versions (version1 & version2)
future challenge is to generate the rest of the version.
Currently 4SQR Code supports only the English language. It is
possible to bring new languages to this system.
This work found the possibility of bringing more states such as
8, 16 & 32 states for QR Codes.
Moreover, 4SQR has an enormous increase in data storage
capacity, but in the main time finding storage capacity, we
had to face the limitation of time and memory compared to
standard QR. This could be minimizable by uncovering a new
and better algorithm.
Conclusion
The usage of the DT in the 4SQR system increases the data stor-
age capacity by 2x as compared to the traditional QR Code. The DT
allows us to increase storage capacity and opens up wider working
domains. Considering the aims of this work, this research’s main
intention was to create a 4SQR-Code based on the traditional QR
Code. Moreover, DT brings us the virtual simulation that opens
up the possible expansion of the 4SQR capacity by 4x, 8x, 16x,
and 32x. It can be noted throughout the work that the proposed
algorithm brings a new and improved version of the QR Code,
and this system can be used more easily and more efficiently in
practical applications. Furthermore, the proposed algorithm is dis-
cussed in several aspects, such as complexity, and data storage
capacity. However, this system needs to polish up more to get
the best output from the system and also can be possible to work
with this system and bring new and improved versions of the QR
Code. Finally, the 4SQR system can generate an extensive number
of QR Codes with increased data storage capacity. Also, after some
future development, the 4SQR system can be the pioneer in the QR
Code field.
Compliance with ethics requirements
This article does not contain any studies with human or animal
subjects
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgment
The authors acknowledge the Researchers Supporting Project
number (RSP2023R34), King Saud University, Riyadh, Saudi Arabia.
References
[1] Scanova. Qr code statistics 2022: Up-to-date numbers on global qr code usage,
[Online; accessed 14-February-2022]; 2022. URL https://scanova.io/blog/qr-
code-statistics/
[2] Wikipedia contributors, Qr code — Wikipedia, the free encyclopedia, https://
en.wikipedia.org/w/index.php?title=QR_code&oldid=1127627244, [Online;
accessed 20-December-2022]; 2022.
[3] Grieves M. Digital twin: manufacturing excellence through virtual factory
replication. White paper 2014;1(2014):1–7.
[4] Singh M, Srivastava R, Fuenmayor E, Kuts V, Qiao Y, Murray N, Devine D.
Applications of digital twin across industries: a review. Appl Sci 2022;12
(11):5727.
Fig. 11. Example output of the proposed system for decoding process.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
15
[5] Huang S, Wang G, Yan Y, Fang X. Blockchain-based data management for
digital twin of product. J Manuf Syst 2020;54:361–71.
[6] Revetria R, Tonelli F, Damiani L, Demartini M, Bisio F, Peruzzo N. A real-time
mechanical structures monitoring system based on digital twin, iot and
augmented reality. 2019 Spring Simulation Conference (SpringSim). IEEE;
2019. p. 1–10.
[7] Wave D. Information technology automatic identification and data capture
techniques qr code bar code symbology specification. Int Organ Standard, ISO/
IEC 2015;18004:23.
[8] Deineko Z, Sotnik S, Lyashenko V. Confidentiality of information when using
qr-coding.
[9] Huang H-C, Chen Y-H, Chang F-C, Lu Y-Y. Enhancements of data hiding
applications with qr code concealment. In: 2021 IEEE 3rd Global Conference on
Life Sciences and Technologies (LifeTech). IEEE; 2021. p. 425–7.
[10] Rahman A, Hossain MS, Muhammad G, Kundu D, Debnath T, Rahman M, Khan
MSI, Tiwari P, Band SS. Federated learning-based ai approaches in smart
healthcare: concepts, taxonomies, challenges and open issues. Clust Comput
2023;26(4):2271–311.
[11] Rahman A, Hasan K, Kundu D, Islam MJ, Debnath T, Band SS, Kumar N. On the
icn-iot with federated learning integration of communication: Concepts,
security-privacy issues, applications, and future perspectives. Future Gener
Comput Syst 2023;138:61–88.
[12] Tan L, Lu Y, Yan X, Liu L, Zhou X. Xor-ed visual secret sharing scheme with
robust and meaningful shadows based on qr codes. Multimedia Tools Appl
2020;79(9):5719–41.
[13] El-Sheref IM, El-Licy FA, Asad AH. Improving performance of the multiplexed
colored qr codes. Int J Adv Comput Sci Appl 11 (3).
[14] Lu Z, Lv W, Zhu Y, Yang D, Zhou X, Wang H, Shi Y. Optical information
encryption based on partially-update iterative system with azimuth
multiplexing. Opt Commun 2022;510:127899.
[15] Yang Y, Lei C, Chen R, Huang H, Tan H-Z. Fast image process and detection of
distorted blur-readable 2d barcode. In: 2022 7th International Conference on
Signal and Image Processing (ICSIP). IEEE; 2022. p. 322–6.
[16] Chen R, Zheng Z, Yu Y, Zhao H, Ren J, Tan H-Z. Fast restoration for out-of-focus
blurred images of qr code with edge prior information via image sensing. IEEE
Sens J 2021;21(16):18222–36.
[17] Ali AM, Farhan AK. Enhancement of qr code capacity by encrypted lossless
compression technology for verification of secure e-document. IEEE Access
2020;8:27448–58.
[18] Arora M, Verma AK, et al. Increase capacity of qr code using compression
technique. 2018 3rd International Conference and Workshops on Recent
Advances and Innovations in Engineering (ICRAIE). IEEE; 2018. p. 1–5.
[19] Baidong H, Yukun Z. Research on quickpass payment terminal application
system based on dynamic qr code, in: Journal of Physics: Conference Series,
Vol. 1168, IOP Publishing; 2019. p. 032059.
[20] Islam MM, Nooruddin S, Karray F, Muhammad G. Multi-level feature fusion for
multimodal human activity recognition in internet of healthcare things.
Inform Fusion 2023;94:17–31.
[21] Rahman A, Islam MJ, Montieri A, Nasir MK, Reza MM, Band SS, Pescape A,
Hasan M, Sookhak M, Mosavi A. Smartblock-sdn: An optimized blockchain-sdn
framework for resource management in iot. IEEE Access 2021;9:28361–76.
[22] Rahman A, Islam MJ, Band SS, Muhammad G, Hasan K, Tiwari P. Towards a
blockchain-sdn-based secure architecture for cloud computing in smart
industrial iot. Digital Commun Networks 2023;9(2):411–21. doi: https://doi.
org/10.1016/j.dcan.2022.11.003.
[23] Islam MM, Nooruddin S, Karray F, Muhammad G. Human activity recognition
using tools of convolutional neural networks: A state of the art review, data
sets, challenges, and future prospects. Comput Biol Med 149 (106060).
[24] Islam MJ, Rahman A, Kabir S, Karim MR, Acharjee UK, Nasir MK, Band SS,
Sookhak M, Wu S. Blockchain-sdn-based energy-aware and distributed secure
architecture for iot in smart cities. IEEE Internet of Things J 2022;9
(5):3850–64. doi: https://doi.org/10.1109/JIOT.2021.3100797.
[25] Rahman A, Islam MJ, Rahman Z, Reza MM, Anwar A, Mahmud MP, Nasir MK,
Noor RM. Distb-condo: Distributed blockchain-based iot-sdn model for smart
condominium. IEEE Access 2020;8:209594–609.
[26] Jiang Y, Yin S, Li K, Luo H, Kaynak O. Industrial applications of digital twins.
Philos Trans Roy Soc A 2021;379(2207):20200360.
[27] Rahman A, Islam J, Kundu D, Karim R, Rahman Z, Band SS, Sookhak M, Tiwari P,
Kumar N. Impacts of blockchain in software-defined internet of things
ecosystem with network function virtualization for smart applications:
Present perspectives and future directions. Int J Commun Syst 2023:e5429.
doi: https://doi.org/10.1002/dac.5429.
[28] Stark R, Fresemann C, Lindow K. Development and operation of digital twins
for technical systems and services. CIRP Ann 2019;68(1):129–32.
[29] Khasanov M, Krasnov F. Digital twin of a research organization: Approaches
and methods. In: SPE Annual Caspian Technical Conference. SPE; 2019. p.
D013S002R004..
[30] Purcell W, Neubauer T. Digital twins in agriculture: A state-of-the-art review.
Smart Agric Technol 2023;3:100094.
[31] van der Valk H, Strobel G, Winkelmann S, Hunker J, Tomczyk M. Supply chains
in the era of digital twins–a review. Proc Comput Sci 2022;204:156–63.
[32] Umaria MM, Jethava G. Enhancing the data storage capacity in qr code using
compression algorithm and achieving security and further data storage
capacity improvement using multiplexing. In: in: 2015 International
Conference on Computational Intelligence and Communication Networks
(CICN). IEEE; 2015. p. 1094–6.
[33] Abas A, Yusof Y, Din R, Azali F, Osman B. Increasing data storage of coloured qr
code using compress, multiplexing and multilayered technique. Bull Electr Eng
Inform 2020;9(6):2555–61.
[34] Wikipedia contributors, Huffman coding — Wikipedia, the free encyclopedia,
[Online; accessed 13-February-2022] (2022). https://en.wikipedia.org/w/
index.php?title=Huffman_coding&oldid=1071647715
[35] Bulan O, Blasinski H, Sharma G. Color qr codes: Increased capacity via per-
channel data encoding and interference cancellation. Color and Imaging
Conference, Vol. 2011. Society for Imaging Science and Technology; 2011. p.
156–9.
[36] Khan MSI, Rahman A, Debnath T, Karim MR, Nasir MK, Band SS, Mosavi A,
Dehzangi I. Accurate brain tumor detection using deep convolutional neural
network, Computational and Structural. Biotechnol J 2022;20:4733–45.
[37] Liu S, Fu Z, Yu B. Rich qr codes with three-layer information using hamming
code. IEEE Access 2019;7:78640–51.
[38] Taveerad N, Vongpradhip S. Development of color qr code for increasing
capacity. In: 2015 11th International Conference on Signal-Image Technology
& Internet-Based Systems (SITIS). IEEE; 2015. p. 645–8.
[39] Galiyawala HJ, Pandya KH. To increase data capacity of qr code using
multiplexing with color coding: An example of embedding speech signal in
qr code. in: 2014 Annual IEEE India Conference (INDICON). IEEE; 2014. p. 1–6.
[40] Querini M, Italiano GF. Color classifiers for 2d color barcodes. In: 2013
Federated conference on computer science and information systems. IEEE;
2013. p. 611–8.
[41] Zheng Z, Yu Y, Chen R, Huang H, Zhao H, Lu X. Localization method based on
multi-qr codes for mobile robots. In: 2022 IEEE International Conference on
Advances in Electrical Engineering and Computer Applications (AEECA). IEEE;
2022. p. 1385–91.
[42] Liu N, Du Y, Xu Y. Qr codes blind deconvolution algorithm based on binary
characteristic and l0 norm minimization. Pattern Recogn Lett
2018;111:117–23.
[43] Wikipedia contributors, Hexadecimal — Wikipedia, the free encyclopedia,
https://en.wikipedia.org/w/index.php?title=Hexadecimal&oldid=1130974725,
[Online; accessed 5-January-2023] (2023).
[44] Querini M, Italiano GF. Reliability and data density in high capacity color
barcodes. Comput Sci Inform Syst 2014;11(4):1595–615.
[45] Melgar MEV, Zaghetto A, Macchiavello B, Nascimento AC. Cqr codes: Colored
quick-response codes. In: in: 2012 IEEE Second International Conference on
Consumer Electronics-Berlin (ICCE-Berlin). IEEE; 2012. p. 321–5.
[46] Gonaboina S, Burra LR, Tumuluru P. Secure qr-pay system with ciphering
techniques in mobile devices. Int J Electron Comput Sci Eng 2012;1
(4):1905–12.
[47] Rahman A, Montieri A, Kundu D, Karim M, Islam M, Umme S, Nascita A,
Pescapé A, et al. On the integration of blockchain and sdn: Overview,
applications, and future perspectives. J Netw Syst Manage 2022;30(4):1–44.
[48] Liu S, Fu Z, Yu B. A two-level qr code scheme based on polynomial secret
sharing. Multimedia Tools Appl 2019;78(15):21291–308.
[49] Fu Z, Cheng Y, Liu S, Yu B. A new two-level information protection scheme
based on visual cryptography and qr code with multiple decryptions.
Measurement 2019;141:267–76.
[50] Cheng Y, Fu Z, Yu B, Shen G. A new two-level qr code with visual cryptography
scheme. Multimedia Tools and Applications 2018;77(16):20629–49.
[51] Lei JTL, Chuin LS, Ernawan F. An image watermarking based on multi-level
authentication for quick response code. In: 2021 International Conference on
Software Engineering & Computer Systems and 4th International Conference
on Computational Science and Information Management (ICSECS-
ICOCSIM). IEEE; 2021. p. 417–22.
[52] Xiong L, Zhong X, Xiong NN, Liu RW. Qr-3s: A high payload qr code secret
sharing system for industrial internet of things in 6g networks. IEEE Trans
Industr Inf 2020;17(10):7213–22.
[53] Wikipedia contributors, Reed–solomon error correction — Wikipedia, the free
encyclopedia, https://en.wikipedia.org/w/index.php?title=Reed%E2%80%
93Solomon_error_correction&oldid=1120074370, [Online; accessed 21-
December-2022] (2022).
[54] Rahman A, Chakraborty C, Anwar A, Karim M, Islam M, Kundu D, Rahman Z,
Band SS, et al. Sdn–iot empowered intelligent framework for industry 4.0
applications during covid-19 pandemic. Cluster Comput 2022;25(4):2351–68.
A.I. Udoy, M.A. Rahaman, Md. Jahidul Islam et al. Journal of Advanced Research xxx (xxxx) xxx
16
