Al-Iraqia Journal for Scientific Engineering Research

Medical images have recently been exposed to unintentional and intentional destruction by unauthorized persons in the storage and transfer of medical images. This undermines the integrity and reliability of diagnostic information. Thus, the paper outlines a watermarking system to authenticate against medical image manipulation. The process involves the application of a sequential chaotic map to encode patient information on a QR code. A texture feature and entropy-based hybrid model are used to determine appropriate image regions to be used in embedding. The system also has an image manipulation detection and locating tool to maintain diagnostic content. X-ray, CT and MRI images were used for experimentation. The findings show reasonable visual quality and data recovery, and achieve high imperceptibility with PSNR value exceeding 56 dB across the different types of attacks, which proves that the proposed method is reliable in ensuring that medical images are not compromised to a range of attacks.

Electrical devices have become an important part of modern daily life. They are widely used in household activities such as cleaning, laundry, refrigeration, cooling systems, and water pumping. The use of these devices increases electricity consumption. A heavy load on the electrical grid is placed during peak hours because of working many electrical devices simultaneously, which results in higher electricity costs due to time-varying energy prices. To solve this problem, scheduling of household electrical devices has become one of the major research topics in energy management systems. Correct scheduling can prevent multiple devices from operating at the same time, which leads to reducing both peak demand and electricity costs by improving overall energy efficiency. This article proposed an enhanced Whale Optimization Algorithm (WOA) to determine the best scheduling for running the household appliances. The proposed approach aims to minimize electricity cost and power consumption while maintaining efficient device operation. In order to improve the exploration capability of the conventional WOA, three chaotic maps are incorporated into the algorithm with a mutation in some places to determine the search positions instead of using random numbers. This modification helps the algorithm converge faster and achieve better optimization performance. The results show that the proposed method provides effective device scheduling and achieves significant reductions in electricity cost and energy consumption compared with the standard WOA approach.

The rapid advancement of deepfake generation techniques poses a significant threat to digital media authenticity, necessitating detection systems that are not only accurate but also explainable and robust across diverse content types. While deep learning models such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have shown promising performance in detecting spatial and temporal inconsistencies, they often operate as black-box systems with limited interpretability and generalization capability. To address these challenges, this paper proposes a neural-symbolic deepfake detection framework that integrates Explainable Artificial Intelligence (XAI) with hybrid deep learning models. The proposed approach combines Region-based Convolutional Neural Networks (RCNN) and Bidirectional Gated Recurrent Units (Bi-GRU) for effective spatiotemporal feature extraction. These features are further processed through a propositional inference layer that incorporates symbolic reasoning based on logical rules reflecting natural facial behavior, including eye movement, lip synchronization, and facial consistency. The model is evaluated on benchmark datasets, including Celeb-DF, Deepfake TIMIT, and WLDR, demonstrating superior performance compared to baseline methods in terms of F1-score, Area Under the Curve (AUC), and Matthews Correlation Coefficient (MCC), along with improved true positive rates in ROC analysis. Furthermore, ablation studies confirm that the integration of symbolic reasoning enhances detection performance by enforcing logical consistency and providing interpretable decision-making. Overall, the results highlight the effectiveness of neural-symbolic reasoning as a robust and transparent framework for deepfake detection, contributing to the advancement of explainable and trustworthy AI systems in multimedia forensics.

The need for bidirectional power flow between Electric Vehicle (EV) chargers and the grid via Vehicle-to-Grid (V2G) technology demands efficient power electronic converters. In modern technology Dual Active Bridge (DAB) has played a pivotal role in the electronic interface (DC–DC) conversion due to its galvanic isolation (High Frequency Transformer (HFT)) structure, bidirectional flow capability, power density, and wide voltage regulation capability. At the core of the efficient operation of the DAB are the Pulse Width Modulation (PWM) techniques. The PWM strategies also extend the zero-voltage switching (ZVS) range and reduce circulating inductive current. This paper reviews and compares the technical aspects of PWM techniques applied in DAB-based EV charging systems, inter alia, Phase Shift Modulation (PSM), Combined PWM (CPWM), Dual Phase Shift (DPS), Triangular and Trapezium Modulation (TTM), PWM plus Phase Shift (PWM–PS), and Asymmetric Duty Cycle Control (ADC). The mathematical analysis of the strategies was done using the power transfer equations and loss mechanisms. Performance evaluation was done based on soft-switching range, circulating inductive current, efficiency, and aptness to Electric Vehicle charging needs. The paper concludes that Dual Phase Shift and PWM plus Phase Shift have superior wide voltage variations, which is standard for fast DC charging, but Phase Shift Modulation is simplified for medium-power onboard chargers.

In this paper, a new solution of inverse kinematics was derived and programmed as an algorithm. The algorithm was coded and embedded in a MATLAB simulation program to manipulate a 5-DOF Humanoid Robotic Arm (HRA) and assess its reaching accuracy. The algorithm was designed as follows: joint frames were modelled based on the Denavit-Hartenberg (D-H) concept. Configuration for each joint frame was designed using the proposed D-H parameters. Forward kinematic (FK) equations were derived using the designed frames configuration. The inverse kinematic problem was solved (derived) using an analytical method to find five joint angles for the desired location and orientation. IK equations were derived from the FK transformation matrix for the given location and orientation to revers back the joint angles. The Graphical User Interface (GUI) was designed using MATLAB to simulate the proposed FK/IK algorithm. A group of desired locations was handled using the GUI to show the resultant pose of the HRA. The accuracy of the proposed IK was assessed by means of positional error. The lowest and highest average positional errors achieved were (0.026 and 0.79) cm, respectively.