Egyptian Informatics Journal最新文献

{"title":"An efficient deep learning system for kinship verification based on ConvNext-EfficientNet-VIT feature fusion","authors":"Nermeen Nader,&nbsp;Fatma El-Zahraa Ahmed El-Gamal,&nbsp;Mohammed Elmogy","doi":"10.1016/j.eij.2025.100809","DOIUrl":"10.1016/j.eij.2025.100809","url":null,"abstract":"<div><div>Kinship verification has emerged as a compelling area of research within computer vision, driven by its critical role in real-world applications, such as forensic investigations and the search for missing persons. Despite recent progress, the task remains challenging due to subtle facial similarities across generations and variations in pose, lighting, and expression. Deep learning techniques have significantly advanced the field. Among them, feature fusion has proven to be a powerful tool for enhancing model performance by integrating complementary characteristics from multiple architectures. This research introduces a new kinship verification framework that harnesses the strengths of ConvNext-Base, EfficientNet-B0, and vision transformer (ViT) through an effective feature fusion strategy. By combining the local texture sensitivity of ConvNeXt, the parameter efficiency of EfficientNet-B0, and the global context modeling capabilities of ViT. The fused representation captures a more holistic and discriminative understanding of facial features relevant to kinship. To the best of the authors’ knowledge, this specific fusion of deep models has not yet been explored for kinship verification. The proposed framework is structured into six stages: image preprocessing, parent/child image pairing, feature extraction, feature normalization, feature fusion, and classification. It was evaluated using two standard benchmark datasets – KinFaceW-I (KinFWI) and KinFaceW-II (KinFWII) – achieving maximum accuracy rates of 84.85% and 91.65%, respectively. These results outperform several state-of-the-art (SOTA) methods and underscore the critical role of multi-model feature fusion in improving the accuracy and robustness of kinship verification systems. This research’s promising findings validate the proposed approach’s effectiveness and highlight the potential of deep feature fusion in addressing complex facial analysis problems.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100809"},"PeriodicalIF":4.3,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cognitive semantic communications in 6G wireless networks using a digital twin approach","authors":"Shahryar Shafique Qureshi ,&nbsp;Muzammil Ahmad Khan ,&nbsp;Muhammad Sheraz ,&nbsp;Teong Chee Chuah ,&nbsp;It Ee Lee ,&nbsp;Muhammad Sajjad","doi":"10.1016/j.eij.2025.100802","DOIUrl":"10.1016/j.eij.2025.100802","url":null,"abstract":"<div><div>This paper introduces an advanced Cognitive Semantic Communication (CSC) framework enhanced through Digital Twin (DT) technology to meet the evolving demands of intelligent and adaptive communication in sixth-generation (6G) wireless networks. Unlike traditional semantic systems that depend on static knowledge representations, the proposed DT-Based CSC incorporates real-time system modeling, dynamic knowledge graphs, and entropy-based reasoning to support predictive optimization and context-aware decision-making. Each communication node is paired with a digital twin that continuously reflects it’s physical, cognitive, and network states, allowing for proactive adjustments based on current conditions. A multi-objective optimization approach is used to balance semantic accuracy, energy efficiency, security, and latency. The framework emphasizes the transmission of meaningful and relevant information rather than raw data, thereby improving overall communication effectiveness. Simulation results confirm that the proposed approach offers substantial improvements over existing methods in terms of semantic precision, system reliability, adaptability, and secure information exchange. The digital twin integration enables the system to maintain stability even under dynamic network conditions, making it particularly well-suited for real-world applications. In conclusion, the DT-Based CSC framework offers a transformative advancement for future 6G communication systems, supporting intelligent services such as autonomous vehicles, smart infrastructure, and decentralized artificial intelligence.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100802"},"PeriodicalIF":4.3,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A routing algorithm for mobile ad-hoc networks using fuzzy logic and hierarchical tree creation","authors":"ZhiYuan Shen ,&nbsp;HaoDe Shen ,&nbsp;WeiFeng Wu","doi":"10.1016/j.eij.2025.100779","DOIUrl":"10.1016/j.eij.2025.100779","url":null,"abstract":"<div><div>Mobile Ad hoc Networks) MANETs) have their limitations because of their dynamic topology, absence of fixed infrastructure, and restricted energy resources. This paper therefore presents a new routing algorithm that incorporates fuzzy logic and hierarchical tree topology to overcome these problems. The first step in the proposed algorithm is to identify the weight of nodes in the network depending on factors such as energy level, number of connections, and memory buffer using a fuzzy inference system. These weights are then used to form clusters and then the formation of a hierarchical tree topology is done. This topology is effective in routing performance. The results of experiments show that the proposed solutions increase the efficiency of a network. The packet delivery ratio was increased by 4% while the energy consumption was reduced by 10.03% and end-to-end delay by 16.08% when the proposed algorithm was used instead of previous methods. This improvement is due to the dynamic weighting of nodes by using fuzzy logic and the stability of the hierarchical tree structure.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100779"},"PeriodicalIF":4.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A machine learning framework for evaluating basketball player skill development using video analysis","authors":"Kai Qin Fang ,&nbsp;Han Jiang","doi":"10.1016/j.eij.2025.100762","DOIUrl":"10.1016/j.eij.2025.100762","url":null,"abstract":"<div><div>A lot of time and effort is used in the conventional approaches to measuring the performance of basketball players. The paper will deal with the necessity of an improved and precise system to assess the skill development of basketball players. The proposed approach uses three-phase process. First, player tracking is done by applying background subtraction and IMM to deal with occlusion problems. The second phase is action recognition by using 3D Convolutional Neural Network (CNN), and finally in the third phase of the proposed method, the player skill is assessed by using ensemble model. This combined model is a new way of automating performance assessment. The model was tested using a set of real NBA match videos and the accuracy of the model was found to be 88.34% for action recognition and 93.19% in evaluating player skills. This goes to show how the proposed approach works well in identifying the various actions of the players and assessing the level of improvement. The proposed framework could be useful for the coaches and analysts to have a better way of evaluating the performance of the players and the training that is to be done. Finally, this framework offers a powerful and effective platform of objective analysis of player performance and training optimization.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100762"},"PeriodicalIF":4.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A recommender system for social networks using link Prediction, clustering and genetic algorithm","authors":"Ming Lu ,&nbsp;Jinglu Chen ,&nbsp;Rongfa Chen ,&nbsp;Xiuzhe Meng","doi":"10.1016/j.eij.2025.100784","DOIUrl":"10.1016/j.eij.2025.100784","url":null,"abstract":"<div><div>In the era of widespread social networks, finding meaningful and relevant connections has become a major challenge. Traditional recommender methods often face challenges in providing accurate and relevant recommendations and cannot fully reflect users’ real interests and connections. These methods are usually unable to provide personalized and efficient recommendations due to the lack of consideration of users’ communication patterns and profile characteristics. Accordingly, in this research, a new recommender method for social networks is presented that operates based on a combination of link prediction, clustering, and genetic algorithm. The proposed method is able to provide more accurate and relevant recommendations by simultaneously considering users’ communication patterns and their profile characteristics. Link prediction helps detect likely relationships between users, while clustering improves prediction ability by clustering users having similar features. Genetic algorithm is also used to determine the best values ​​of the model’s most significant parameters. Thus, the model is able to dynamically adjust to the data and provide the best performance in a number of dimensions such as precision, recall, and F-measure. Experimental findings revealed that the proposed method outperformed the conventional methods and achieved over 98% accuracy. As such, this hybrid technique, through the use of efficient clustering, parameter tuning, and proper link prediction, has been able to act as a good recommender system for social networks and fulfills the need to detect meaningful and valuable connections in a positive manner.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100784"},"PeriodicalIF":4.3,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A robust deep learning pipeline for multi-class cervical cancer cell identification","authors":"Entesar Hamed I. Eliwa ,&nbsp;Tarek Abd El-Hafeez","doi":"10.1016/j.eij.2025.100787","DOIUrl":"10.1016/j.eij.2025.100787","url":null,"abstract":"<div><div>Cervical cancer remains a significant global health concern, necessitating accurate and early diagnostic tools. This paper presents a robust deep learning pipeline for the multi-class classification of cervical cancer cells, introducing an optimized YOLO-based architecture enhanced with a novel Attention-Guided Multi-Scale Feature Fusion (AGMS-FF) module. We conduct a comprehensive evaluation, comparing our Full AGMS-FF model against its Baseline YOLOv11 counterpart and established convolutional neural networks including EfficientNet-B0, MobileNetV3, and ResNet18. Experiments were rigorously conducted on two distinct cervical cell datasets to assess model robustness and generalization. On Dataset 1 (9,500 images), the Full AGMS-FF model achieved the highest accuracy of 0.9256 and an exceptional Macro AUC of 0.9910, outperforming EfficientNet-B0 (0.8330 accuracy), MobileNetV3 (0.8028), and ResNet18 (0.7324). The Baseline YOLOv11 also demonstrated strong performance (0.9235 accuracy, 0.9910 Macro AUC). On the more challenging Dataset 2 (4,966 images), the Full AGMS-FF model again led with an accuracy of 0.8471 and a Macro AUC of 0.9791, with Baseline YOLOv11 at 0.8310 accuracy and 0.9774 Macro AUC. Both YOLO-based models consistently demonstrated remarkable performance across a wide spectrum of cervical cell classes, showing particularly high F1-scores for both benign and certain abnormal cell types. In contrast, other CNNs struggled notably with less frequent or more challenging pathological variations. Our findings highlight the profound capability of the enhanced YOLO architecture in achieving high-precision, multi-class classification across different data distributions, offering a promising avenue for clinical diagnostic support.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100787"},"PeriodicalIF":4.3,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing wireless sensor network performance through self-tuned fuzzy logic, adaptive palm tree optimization, and Stackelberg Game-Theoretic load balancing: A comprehensive approach for energy efficiency, reliability, and security","authors":"Faridha Banu D. ,&nbsp;Kumaresan N.","doi":"10.1016/j.eij.2025.100767","DOIUrl":"10.1016/j.eij.2025.100767","url":null,"abstract":"<div><div>A wireless sensor network (WSN) is a network of geographically scattered sensors that collects and transmits environmental data to a central processing station. Despite their extensive usage in environmental monitoring, military surveillance, and healthcare, WSNs present considerable challenges, including energy efficiency, network durability, and data transfer reliability. High energy consumption, irregular data throughput, packet loss, and lower network lifetime have detrimental effects on WSN performance, particularly in dynamic and large-scale configurations. This work introduces an innovative three-layer framework aimed at improving clustering, routing, and load balancing in Wireless Sensor Networks (WSNs). Initially, dynamic clustering is accomplished through a combination of Self-Tuned Fuzzy Logic and Adaptive Palm Tree Optimization (APTO), which considers energy, distance, throughput, and trust to effectively choose Cluster Heads (CHs). Next, the improved orbit optimization algorithm (IOOA) is utilized for selecting multi-hop routing paths and optimizing factors, such as energy usage, latency, and reliability. Finally, a Stackelberg Game-theoretic Approach (SGTA) is applied to ensure a fair load distribution among nodes, preventing overload, and boosting network stability. Together, these methods enhance the energy efficiency, reliability, and overall performance of the WSN. Simulation results demonstrate that the proposed approach, compared to existing algorithms such as the game-based dynamic clustering routing (GDCR) protocol, Game Theory-Based Fuzzy Routing Protocol (GTFR), Energy and Throughput Aware Adaptive Routing (ETAAR) algorithm based on Cooperative Game Theory (CGT), game theory inter-cluster routing, improved ant colony optimization (GTIACO), enhancement game, and gray wolf algorithm (EG-GWA) protocol, decreases energy consumption, improves throughput, extends network lifespan, and provides stability and reliability. The proposed method achieves an energy consumption of 15 mJ, packet delivery ratio (PDR), 0.98 Mbps of throughput, and 0.32 ms of jitter.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100767"},"PeriodicalIF":4.3,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Proposal to strength image encryption using blockchain and hybrid chaotic-DNA techniques","authors":"Sanaa A. Jabber ,&nbsp;Ayad Al-Adhami ,&nbsp;Rajaa K. Hasoun ,&nbsp;Rasha S. Ali ,&nbsp;Soukaena H. Hashem","doi":"10.1016/j.eij.2025.100765","DOIUrl":"10.1016/j.eij.2025.100765","url":null,"abstract":"<div><div>This research proposes a novel image encryption framework that synergizes blockchain technology with a hybrid approach combining chaotic systems and DNA-based encoding. The presented method begins by decomposing a color image into its red, green, and blue channels. Each channel is then converted from binary to DNA sequences using a predefined mapping, followed by algebraic DNA operations to further obscure the data. The DNA-encoded image is reconstructed and subjected to the Rossler chaotic system, which, using certain initial values and constants, produces three normalized chaotic sequences. These sequences are generated numerically via the Runge–Kutta method. The chaotic sequences are then utilized to permute the image’s rows and columns and to generate additional keys for DNA-based XOR operations. After encryption, the SHA-256 hash of the encrypted image is computed, and the image is uploaded to the Inter Planetary File System (IPFS) to obtain a unique content identifier. All relevant metadata, including the content identifier, hash value, encryption keys, and parameters, are securely stored on a blockchain through a smart contract, ensuring data integrity and non-repudiation. The decryption process reverses these steps to restore the original image. Experimental results confirm the system’s effectiveness, demonstrating high distortion between original and encrypted images and strong resistance to various cryptanalytic attacks. Experimental results show that; MSE values (ranging roughly between 950 and 1300) and PSNR values (approximately 13–15 dB) and some of other measurements indicate a high level of distortion between the original and encrypted images.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100765"},"PeriodicalIF":4.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reservoir dam monitoring technology by integrating improved ABC algorithm and SVM algorithm","authors":"Xingang Wang ,&nbsp;Zhongbo Liu ,&nbsp;Yang Zhao","doi":"10.1016/j.eij.2025.100783","DOIUrl":"10.1016/j.eij.2025.100783","url":null,"abstract":"<div><div>The safe operation of reservoir dams is crucial for the development of human society and economy, but they are easily deformed due to various factors such as climate and water flow, posing a threat to their safe operation. In response to the above issues, this study proposes a reservoir dam deformation monitoring technology based on artificial bee colony algorithm and least squares support vector algorithm. This study proposes optimization strategies to improve the parameter optimization ability of traditional artificial bee colony algorithms by addressing their shortcomings such as randomness, susceptibility to local optima, and insufficient exploration capabilities. On this basis, data preprocessing operations such as singular value removal, Lagrangian interpolation, and wavelet denoising are carried out on the deformation monitoring data of the reservoir dam. The experiment showed that the optimization algorithm achieved optimal and worst values of 0.00E + 00 on the Rastigin and Ackley functions. The maximum absolute deviation of the proposed model was 0.537 mm, the minimum deviation was −0.017 mm, and the maximum and minimum relative errors were 11.26 % and 0.45 %. Comparative verification showed that the MAE, MAPE, and RMSE values of the proposed model were 0.189, 4.82, and 0.256, respectively, which were better than the comparison algorithms.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100783"},"PeriodicalIF":4.3,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145107958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing personalization in IoT-based health monitoring via generative AI and transfer learning","authors":"Rupali Atul Mahajan ,&nbsp;Rajesh Dey ,&nbsp;Mudassir Khan ,&nbsp;Mazliham Mohd Su’ud ,&nbsp;Muhammad Mansoor Alam ,&nbsp;Pratibha Jadhav","doi":"10.1016/j.eij.2025.100788","DOIUrl":"10.1016/j.eij.2025.100788","url":null,"abstract":"<div><div>Owing to the rapid expansion of Internet of Things (IoT) devices, the health care sector is responsible for immense amounts of real-time data, which provides an impetus for custom health metrics. In this context, the current research seeks to fill this gap by proposing a groundbreaking system that employs generative AI technologies and transfer learning in the field of IoT-based health monitoring. Before examining the IoT health data, we must remove any potential discrepancies and errors through data cleaning. An adaptive filter referred to as the delayed error normalized LMS (DENLMS) is a highly sophisticated method that essentially contributes to increasing the precision and accuracy of these particular data. By applying analysis in the frequency domain to the data, we were able to extract features via the fast Fourier transform (FFT) and subsequently review sessions that contained, for example, heart rate variability or respiratory signals over time. The process of developing a generative AI model for personal health monitoring involves selecting suitable models, such as generative adversarial networks (GANs) or variational autoencoders (VAEs), owing to their ability to generate and simulate health data patterns effectively. To facilitate functional data analysis, the system design integrates machine learning techniques with generative models for patient data from various IoT devices. Importantly, the accuracy rate of this technique is 95.6%, the precision rate is 96.4%, the recall rate is 94.7%, and the F1 score is 95.5%. These metrics surpass those of most other techniques described in this study, demonstrating the superior performance of this research technique over other generic algorithms and its implementation with Python software. Future research could also focus on addressing the seemingly trivial challenge of enhancing model adaptability and scalability to meet individual health requirements and integrate multiple data sources.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"32 ","pages":"Article 100788"},"PeriodicalIF":4.3,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145107930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}