IRAQI JOURNAL OF COMPUTERS, COMMUNICATIONS, CONTROL AND SYSTEMS ENGINEERING

The traffic surveillance system is a type of intelligent system of
traffic control. Traffic control provides solutions to most problems faced by
people. It helps to monitor, detect traffic congestion and traffic accidents. As
science evolved, it became possible to control traffic using video surveillance.
Video surveillance is the most economical option that does not involve high
costs or changes in infrastructure. Vehicle detection is one of the main parts of
the traffic surveillance system. In this paper, vehicles will be detected using two
different artificial intelligence methods (the YOLO method and the HAAR
cascade classifier method). The first one is smarter than the second method, and
both of them contain machine learning. The first processing step will read the
video. Then vehicle detection algorithms are applied using two different ways.
The comparison between them depends on the results to find the most effective
and applicable vehicle detection method. After implementing the two methods,
results were obtained using YOLO, that the accuracy is 91.31% and the error
rate is 8.69%, in time 10 sec. As for using the XML (HAAR cascade classifier
method) method, the accuracy is 86.9%, the quality is 86.9%, completeness is
90.9%, and the error rate is 13%, in time 17 sec. Thus, we conclude that the
YOLO method has better results than the second method.

Alzheimer’s disease (AD) is caused by multiple variables.
Alzheimer's disease development and progression are influenced by genetic
variants. The molecular pathways causing Alzheimer's disease are still poorly
understood. In Alzheimer's disease research, determining an effective and
reliable diagnosis remains a major difficulty, particularly in the early stages
(i.e., Moderate Cognitive Impairment (MCI)). Researchers and technologists
working in the fields of machine learning and data mining can help improve the
situation, early AD diagnosis but face a hurdle when it comes to high-
dimensional data processing. By reducing irrelevant and redundant data from
microarray gene expression data, the technique of feature selection can save
computing time, improve learning accuracy, and encourage a deeper effect on
the learning system or data. The feature selection strategy described in this
article reduces data noise well. In particular, Pearson's correlation coefficient
is used to assess data redundancy. The efficacy of these features is assessed
using the Support Vector Machine (SVM) classification approach. The proposed
approach has an accuracy of up to 91.1 %. As a result, newly established
approaches for early diagnosis of Alzheimer's disease(AD) are being improved.

Cybersecurity systems have been taken into account in modern
information systems and methods. This is due to the increase in electronic
attacks on storage information in terms of transmission, reception and storage.
Therefore, the need to produce such systems in a complete way to prevent their
penetration into the network has increased. In addition, artificial intelligence
(AI) methods are used in cybersecurity systems as classifiers, attack detectors,
and components for predicting threats that surround the network. This requires
more information about threats and vulnerabilities to be covered to avoid any
errors. In this paper, a systematic review is conducted to cover cybersecurity
used in various applications, including systems based on wireless, cloud, and
mobile sensor networks (WSN). The systematic review approach is adopted on a
two-way basis to produce a clear view of the research work to date and to
provide a field that can be used for future work.

Wireless Sensor Network (WSN) represents a key network in the
present and future Internet of Things (IoT) technology. WSN has an
uncountable number of applications and is commonly used to aggregate
information and control the physical environment remotely through small
embedded devices known as wireless sensor nodes. Power consumption is one
of the main challenges in WSN due to the limitation of power resources.
Consequently, several techniques have been followed to optimize power
consumption. The feedback control system is one of the routes that has been
utilized to minimize power consumption in WSN using the mathematical model
of power and rate control in WSN. In this paper, a concise review of various
types of control systems that are deployed for power saving in WSN will be
discussed. The comparison between the applied control strategies is the key
finding.

Computed tomography (CT) is used to diagnose diseases and
tumors. A special dye called contrast material is used in CT scans to assist
emphasize the parts of the body being examined. Therefore, an enhancement
technique to improve CT images' degradation is needed. This paper aims to
present a method to enhance the quality of Ct-Scan images based on discrete
wavelets transform and the Retinex algorithm. The proposed methods are based
on the Retinex algorithm parameters or Dark Channel Prior algorithm
parameters, according to the output image from preprocessing and discrete
wavelets algorithm steps to increase the lightness degree of an image, remove
possible noise from the image improve the contrast. The results of experiments
of the enhanced image outputted from the Retinex model compared with one
outputted from the Dark Channel Prior method. Hence, image quality based on
the DCP method is a higher degree of enhanced rate and reasonable rate of
noise removal-based enhancement measures, which were SI, MSE, IQI, SNR,
and SNR, but was very attentive to the percentage values of IQI and SI. So the
DCP with WT was recorded as the highest rate of enhancement.

Testing is a vital phase in software development, and having the
right amount of test data is an important aspect in speeding up the process. As a
result of the integrationist optimization challenge, extensive testing may not
always be practicable. There is also a shortage of resources, expenses, and
schedules that impede the testing process. One way to explain combinational
testing (CT) is as a basic strategy for creating new test cases. CT has been
discussed by several scholars while establishing alternative tactics depending
on the interactions between parameters. Thus, an investigation into current CT
methods was started in order to better understand their capabilities and
limitations. In this study, 97 publications were evaluated based on a variety of
criteria, including the generation technology, test strategy method, supported
interactions, mixed coverage ,and support constraints between parameters. CT
analysis had a wide range of interaction assistance options available to
researchers. Since 2010, a unified interaction has been the most common style
of interaction between the two parties. The year 2018 was hailed as the most
successful in terms of CT by researchers. Researchers should focus on one test
at a time and metaheuristic search strategies for t-way CT. There has also been
a significant increase in the popularity of other trends, such as deep learning
(DL). CT appears to be a useful testing technique for balancing and fault
detection capabilities for a variety of systems and applications, according to
our research. Future research and software development may benefit from this
information.

Four Fractional/Integer Order Fuzzy Proportional Integral
Derivative controller structures are designed in this study to successfully
control a nonlinear, coupled, multi-input, multi-output, three-link rigid robotic
manipulator system. The performance of Fractional Order Fuzzy Proportional
Integral Derivative and Integer Order Fuzzy Proportional Integral Derivative
controllers is evaluated for reference trajectory tracking, changing beginning
circumstances, disturbance rejection, and model uncertainty. These controllers'
parameters are tuned using a meta-heuristic optimization approach called the
most valuable player algorithm for the objective function, which is defined as
the integral of the time-squared error. Simulation results show that the
suggested Fractional Order Fuzzy Proportional Integral Derivative controllers
outperform Integer Order Fuzzy Proportional Integral Derivative controllers
for tracking performance, stability, and robustness for all structures. Fractional
Order Fuzzy Proportional Derivative Fractional Order Proportional Integral
Derivative controller is the best one for trajectory tracking, disturbances
rejection, and parameter variation with the least integral of time square error
equal to 2.7420×10-6, 3.4×10-3 and 2.0108×10-4 respectively and the response
of the angular position for all links for trajectory tracking has minimum settling
time which is equal to 0.0290 s for the first link, 0.0160 s for the second link and
0.0050 s for the third link. When the initial condition is changed, the One Block
Fractional Order Fuzzy Proportional Integral Derivative controller is the best
one, since the integral of time square error is minimum and equal to
1.6253×10-4
.

The search algorithms are characterized by their ability to find the
optimal path in a short calculation time. In this study, a comparative analysis
has been conducted to perform path planning of planar manipulator for static
obstacle avoidance based on graph search algorithms. Four methods have been
taken into account to establish a comparison platform; namely, conventional
A*, modified A*, Chaos A*, and circulation heuristic search (CHS) algorithms.
The performance of comparison is evaluated in terms of length of optimal path
and consumption time of calculation. All algorithms have been coded and
simulated within the MATLAB software environment. According to computer
simulation, the results showed that CHS algorithms outperform the other graph
search ones in terms of generated path length, while the Choas A* could give
the least calculation time as compared to its counterparts.

The Digital watermarking is a field of information hiding that
entails hiding the crucial information in the original data in order to prevent
illegal duplication and distribution of multimedia data such as image, video,
text and ect.. In this paper, we present two techniques to embed watermarks in
the cover image. The first is the Least Significant Bit (LSB) method, which is a
spatial domain technique and considered fragile against attacks and other
operations. The second method is the frequency domain technique, which uses
Discrete Wavelet Transform (DWT) and is considered robust against attacks.
The efficiency and performance of these techniques are evaluated based on
Peak Signal to Noise Ratio (PSNR) and Mean Square Error (MSE). From the
results, the value of PSNR is above 37 dB, which ensures better imperceptibility
and shows better robustness. The comparison between the two techniques shows
that the hybrid method was more robust than the LSB method, hence it achieves
good invisibility.

the primary concerns with manual transactions include corruption, lack of
transparency, fraud, and mismanagement of distribution operations, all of which are
created by traditional centralized applications, necessitating the migration to blockchain
technology. In this work, a system is presented to secure and monitor correspondence
between several nodes and store it in a decentralized database in order to secure
distributed ledger transactions and safeguard against fraud and tampering when
transactions are shared by multiple parties. The hashing that blockchain technology
delivers in each transaction ensures a high level of security. The hashing associated with
each transaction confirms all sending and receiving transactions. When a transaction is
sent from one node to another, the other node checks the hash accompanying the
transaction to see if it came from a registered node or an external node. Within the
blockchain system, the nodes will check transaction correspondences. The system has
demonstrated its effectiveness by delivering a more secure messaging system with high
credibility and tamper resistance. In addition, the time it takes to authenticate will be in
real time.

Human gait data follows distinct and identifiable patterns that are critical for
movement analysis and evaluation Like other biological signals. The success of a
rehabilitation program is dependent on the execution of proper progress monitoring. To
ensure success, diagnosis of gait anomalies, as well as the implementation of therapy to
address them, must be validated in a constant and timely manner in developing youngsters.
In this paper, machine learning techniques were utilized to classify foot diseases and the
purpose is to increase the accuracy of disease detection and diagnosis because intelligent
systems can contribute significantly in the medical field and have proven their worth in
diagnosing many diseases. The results show high accuracy of the used machine learning
algorithms, where the accuracy of the classifiers reached 100% for Random Forest (RF),
Decision Tre

The rapid growth of the human population and the spread of many bad habits
affect the health of human beings, this leads to many health problems, such as heart disease,
blood pressure, and diabetes. Some of these diseases require earlier detection and fast
treatment, to avoid major risks, such as permanent damage or even life loss. The fast
evolution of communications and nanotechnology nowadays is being facilitated to help with
saving lives before great damage happens. This paper suggests a special architecture
through which to monitor patients remotely using an Internet of Things (IoT)architecture for
the purpose of notifying the paramedics and health care centers to rescue the life of
cardiovascular patients. The experiments, after attempts with several types of classification
algorithms, showed that The result shows that The decision tree has high accuracy for
predicting the new heart attack and saving lives before happening of great damage .the result
reached between( 83 to 87) percentage, which is a good percentage to avoid the risk of heart
attacks and thus reducing death rates.

Image Protection is one of the most important issues that have created
problems in technology challenges in the past and present years, whether they are stored
or when sent to other parties. And how to develop the techniques adopted in encrypting it,
devise new methods, or integrate the available technologies to provide high security in
encrypting images or any important data and preserving them from loss or hacking. This
paper presents a new method to generate a random keychain using genetic algorithm
techniques to develop new generations and XOR technology to encrypt digital images.
Where the results showed the high efficiency of the encryption method with its ease of use
and the preservation of the original image data after decoding with high accuracy and
speed in implementing the strategy used. The efficiency of these switches for use has been
tested using the National Institute of Standards and Technology (NIST) and the statistical
randomness test, and the tests were successfully passed.

The most attractive study framework among academics is Software Define
Networking Networking SDN, which aims to create the Internet with an architecture-
independent architecture that will lead to the most significant advances in the network field.
This can solve many network problems to deal with high demand changes and reduce
Replenishment, changes and less manual work. Because of the limited architecture of
traditional networks, which requires modifications in the basic design, network expansion
has been mature and slow. Since 2010, until now, the ODL - OpenDayLight model has been
proposed to solve most of the problems that guide network engineers in the process of
managing complex high-volume networks by top research-oriented universities around the
world. Now is the time to turn dreams into reality by putting the presented ideas into action,
which will result in a solution that meets the expectations of the researcher regarding the
process of managing complex networks and all forms of networks. This document is an
attempt to assist researchers in implementing a software identification network
infrastructure on which the research community may focus on further analysis and
development. We demonstrated an incremental approach to implementing ODL -
OpenDayLight Controller (ODL is a JVM program and can run from any operating system
and device as long as it supports Java) from the Software Define Network, as well as creating
and executing required scenarios, and illustrate the working nature of ODL - OpenDayLight
compared to ryu contrlloer, in this paper Research.

Honeywords are fake passwords that are typically companions to
the real password “sugarword.” The honeyword technique is a password
cracking detection technique that works effectively to improve the security of
hashed passwords by making password cracking simpler to detect. The
password database will contain many honeywords for each user in the system.
A silent alarm will trigger, indicating that the password database has been
compromised if the hacker signs in using a honeyword. The honeychecker is a
separate server in charge of recognizing the real password and raising the
silent alarm. Many honeyword creation techniques have been presented
previously. They all have limitations in the generating process, supporting
characteristics, and strengths of honeyword. The bees algorithm, an
optimization metaheuristic swarm intelligence algorithm, is used in this article
to suggest a novel approach for generating honeywords. The proposed bee
algorithm succeeded in addressing the limitations of the previous methods by
enhancing the honeyword generating process, supporting the honeyword
characteristics, and addressing the honeyword system problems. The most
important characteristics of the honeyword (flatness, DoS resistance, and
storage) were supported by the proposed method to present unconditionally
flatness, strong DoS resistance, and moderate storage.