Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

31,377 Article Results

URL-based phishing detection using XGBoost with engineered features

10.11591/ijeecs.v43.i3.pp908-927
Jawaher Alharbi , Manal Bayousef , Hind Almisbahi
URL-based phishing involves fake uniform resource locators (URLs) created by attackers to trick users into believing they are visiting a legitimate website and thereby steal their confidential information. While several powerful machine learning (ML) and deep learning (DL) studies exist to detect phishing, they still face limitations. Many studies rely on third-party intervention to extract features, which introduces delays that make them unsuitable for fast detection. Another limitation is that existing studies often use small datasets, and traditional features hinder models' ability to learn new phishing techniques, resulting in poor generalization. Therefore, developing new features is crucial to ensure that anti-phishing tools can keep pace with evolving phishing tactics. In addition, the existing studies do not report detection time, which is important for fast detection, and reduces methodological clarity. This paper aims to address these limitations by applying a neural network model and traditional ML classification algorithms to support browser-based phishing detection that balances high accuracy with fast detection. Our XGBoost model achieved 98% accuracy on the test set, utilizing 40 third-party-independent features. Additionally, we achieved an average response time of 0.026225 seconds and an average computation time of 1.6977×10⁻6 seconds per URL, which demonstrates competitive speed. We provided a table of features from recent studies, together with their documented sources, to support future research and analyze key URL-based features.
Volume: 43
Issue: 3
Page: 908-927
Publish at: 2026-09-01

Design and performance optimization of a magnetic based fault current limiter using BBO

10.11591/ijeecs.v43.i3.pp695-708
Tirtha Sankar Daphadar , Amalendu Bikash Choudhury , Tapan Santra
Magnetic fault current limiters (MFCLs) are taking on greater importance in the modern power system because of their ability to suppress excessive fault currents, while having no effect on normal operation. This paper proposes a simulation-based design and performance optimization of a magnetic-based fault current limiter by using biogeography-based optimization (BBO). A coupled electromagnetic-electrical model is developed to describe the nonlinear interaction between current, inductance, magnetic flux, core saturation, and magnetic biasing. The optimization is concentrated on the core geometry, length of the air gap, number of turns of the winding, and size of the magnetic element to ensure effective reduction of the fault current without violating voltage and thermal limits. Finite element analysis is used to study magnetic field distribution, saturation phenomenon, and temperature rise in normal and faulty conditions. Time-domain simulations are also performed for symmetrical and asymmetrical fault cases for transient performance evaluation. The optimized MFCL shows a great reduction of peak fault current and energy dissipation compared with the initial design, with negligible impact during normal operation. Sensitivity and parametric analyses validate the optimization design robustness, which shows that the proposed BBO-based optimization design framework is effective in improving MFCL performance in power system protection applications.
Volume: 43
Issue: 3
Page: 695-708
Publish at: 2026-09-01

Visual and electrical approaches for automated verification of electronic components

10.11591/ijra.v15i3.pp658-668
Sowmya Santhanam , Aadhitya Swaminathan Velmurugan , Durkadevi Chandrahasan , Krithika Arcot Rathna Kumar , Jeevanthika Chepauk , Deepika SasiKumar , Chaithanya Sarangaraj
Reliable inspection and sorting of electronic components have become pivotal along the path of electronic manufacturing toward higher density and automation. Automated optical inspection systems at present depend on visual assessment methods because they do not include electrical testing capabilities, which results in a component verification reliability gap. In spite of recent advances, largely due to the fact that most automated optical inspection systems still abide by visual evaluation, verification of the actual electrical behavior of components became quite impossible. This paper is focused on bridging this gap through the introduction of a unified inspection framework whereby visual analysis is executed along with programmable electrical validation under a single automated process in conformity with Industry 4.0 practices. The system's synchronized workflow includes vision-based detection, optical character recognition, resistor color-band parsing, surface defect analysis, and electrical testing in real time. The component localization task uses YOLOv5, while EasyOCR with a convolutional neural network-long short-term memory (CNN-LSTM) structure and HSV-based segmentation delivers exact value extraction results. The testing system achieved 98.4% classification accuracy, 98.9% value recognition accuracy, and 94.9% overall sorting accuracy when tested on 3,000 photos and 400 physically inspected components at a throughput rate of seven components per minute.
Volume: 15
Issue: 3
Page: 658-668
Publish at: 2026-09-01

ROVAA: Offline attendance automation using a voice–OCR-based 3-DOF robotic arm with Raspberry Pi

10.11591/ijra.v15i3.pp561-576
Rajanikanth Kashi Nagaraj , Archana Harihara Ranganatha , Surendra Hanumanthaiah Honnamachanahalli , Venu Manighatta Gopalakrishnappa
Conventional classroom attendance systems suffer from limitations in accuracy, hygiene, data privacy, and reliability in low-connectivity environments, whether they are manual, cloud-dependent, or single-modality systems. To address these gaps, this paper presents ROVAA, a low-cost, fully offline, AI-driven robotic attendance system in the classroom environment that uniquely integrates three complementary modalities: offline voice recognition, optical character recognition (OCR), and a 3- degrees of freedom (DOF) robotic arm controlled via inverse kinematics, an integration not demonstrated in prior work. The system operates on a Raspberry Pi 4 model B and employs the Vosk speech recognition model and Tesseract OCR for accurate offline processing. Audio and visual inputs are matched in real time to enable the arm to mark attendance at pre-calibrated positions on a touchscreen. Experimental validation under varied lighting and acoustic conditions yielded 96.2% speech recognition accuracy, 95.8% OCR accuracy, and 97.6% robotic arm precision, producing an overall system success rate of 92.8%, demonstrating that high reliability is achievable without cloud infrastructure. The system is designed for cost-effectiveness, data privacy, and scalability, making it suitable for resource-constrained environments such as rural schools and institutions with limited network access. It additionally serves as an educational platform for human–robot collaboration.
Volume: 15
Issue: 3
Page: 561-576
Publish at: 2026-09-01

Optimal robust control for self-balancing robot-based feedback linearization and Atom Search Optimization

10.11591/ijra.v15i3.pp519-528
Alaa Jumaah Al-Maiahy , Yahya Ghufran Khidhir , Adnan Jabbar Attiya , Hisham H. Jasim
In this paper, a robust control method is suggested for attitude control of the two-wheeled self-balancing robot by combining feedback linearization with sliding mode control techniques. The proposed method takes into account important challenges such as external disturbance and system uncertainty. Feedback linearization cancels the nonlinearities in the dynamics of the robotic system, while the sliding mode control handles the uncertainties and the external disturbance. The parameters of the proposed controller are selected by tuning the controller with the Atom Search Optimization algorithm. MATLAB is used to simulate the proposed controller. Simulation results indicate a good performance of the presented controller with high robustness compared with the proportional-integral-derivative (PID) controller. Moreover, the proposed method reduces the rise time by approximately 40% and 50% with respect to PID. These results illustrate the feasibility of the presented control method to be used for real-time implementation in autonomous robotic balancing systems.
Volume: 15
Issue: 3
Page: 519-528
Publish at: 2026-09-01

IoT-based approach to locate books in library system for integration with the automated library systems

10.11591/ijra.v15i3.pp647-657
Sagar S. Ajanalkar , Harshadeep S. Joshi
The main challenge in a large library is to retrieve a book effectively, where traditional barcode- or RFID-based systems require significant infrastructure and complex inventory management. An integrated IoT system consisting of LED-assisted shelf identification and an interactive web visualization application is proposed in this study for enhancing the reliability and effectiveness of automated library book retrieval. In this, a networked microcontroller architecture is used for real-time identification of the exact physical location of books by allowing users to visually locate the required books to be retrieved using LED indicators. Along with this, an interactive web visualization application is developed using HTML5, CSS, JavaScript, PHP, JSON, and MySQL for visualization of the shelf real-time location and location of the book within the shelf. The proposed system offers a cost-effective, scalable, and user-friendly solution compared to conventional RFID or barcode-based systems. Experimental testing of the proposed system passed all 9 tests conducted for its conformance in identifying the required book locations across four distinct test conditions. Beyond the library book retrieval application, the proposed system architecture has potential applications in smart warehouses, storage inventory management, and application as a supporting architecture for future systems such as autonomous service robots.
Volume: 15
Issue: 3
Page: 647-657
Publish at: 2026-09-01

AI driven automated library assistance using pick-to-light system

10.11591/ijra.v15i3.pp639-646
Archana S. Ubale , Vaishali Baste , Harshada Bhushan Magar , Nilakshee R. Rajule
This paper presents the design and implementation of a library assistance system that uses a pick-to-light mechanism and integrates AI-based book recommendations. Book retrieval done manually in libraries is frequently time-consuming, prone to errors, and inefficient. As a solution, we will suggest a hybrid library assistance system, which would be based on an ATmega328 microcontroller, user identification by RFID, pick-to-light, and an AI-based collaborative filtering recommender. The user is directed to the preferred books through the LEDs placed on the shelves, and the AI module gives the user personalized suggestions. The experimental outcomes of 100 users show significant advances over the old traditional manual processes. The time spent searching for a book dropped by 96.4 seconds to 34.8 seconds, the error rate dropped to 3.2%, user satisfaction went up to 4.6, and the success rate of the entire book search process went up to 98.1%. The 500 retrieval cycles of the stress testing showed stability, reliability, and uniformity of the system, and the performance of the LED-RFID. These findings suggest that the proposed system is highly efficient for the library, minimizes human error, and positively impacts the user experience, making it applicable to real-life library environments.
Volume: 15
Issue: 3
Page: 639-646
Publish at: 2026-09-01

Fall detection system based on accelerometer/gyroscope data fusion

10.11591/ijeecs.v43.i3.pp726-736
Amina Makhlouf , Isma Boudouane , Nadia Saadia , Amar Ramdane-Cherif
Falls are one of the major causes of death among the elderly worldwide. Faced with this public health challenge, many researchers have designed continuous monitoring systems to detect this type of incident at an earlier time, in order to facilitate immediate treatment and reduce potentially serious consequences. In this paper, we present a solution for the continuous monitoring of elderly or disabled people. We developed a fall detection system based on the fusion of data from a gyroscope and an accelerometer of a wearable inertial station. Two algorithms were developed: the first uses only the data from the gyroscope, while the second is based on the fusion of gyroscopic and accelerometric data. The evaluation of the system took into account several types of falls and activities, as well as different categories of people of different gender, age, height and weight, taken from the SISFALL database. The algorithm based on the gyroscope achieved an accuracy of 87.13% and a specificity of 84.27%, while its fusion with the accelerometer improved these performances to 90.76% and 89.92%, respectively, demonstrating the benefit of sensor fusion.
Volume: 43
Issue: 3
Page: 726-736
Publish at: 2026-09-01

Zeroing neurodynamics proportional-integral-derivative controller for order-2 system: design, analysis, and verification

10.11591/ijra.v15i3.pp690-697
Yunong Zhang , Junyan Liu , Zhonghua Li
This paper presents a zeroing neurodynamics proportional-integral-derivative (PID) controller for an order-2 single-input single-output (SISO) affine-in-control nonlinear system. The controller is derived from zeroing neurodynamics (ZN) and reformulates PID gains through ZN parameters, establishing an explicit correspondence to the closed-loop poles of the error dynamics. This enables systematic pole placement for stability and performance tuning, eliminating empirical trial-and-error. Theoretical analysis proves exponential convergence of the tracking error. The inverted pendulum, a benchmark system with strong geometric nonlinearity, is used to verify the controller. Simulations with large initial offsets and constant references show that the proposed controller maintains stable tracking in operating conditions where the conventional PID controller exhibits substantial performance degradation. The resulting framework provides a model-based and theoretically grounded alternative for nonlinear control.
Volume: 15
Issue: 3
Page: 690-697
Publish at: 2026-09-01

Cost-aware global frontier matching for ROS 2 multi-robot exploration

10.11591/ijra.v15i3.pp589-596
Chu Van Cuong , Tran Tuan Anh
Multi-robot frontier exploration supports warehouse mapping and inspection robotics, but geometric assignment can produce overlapping motion and inefficient target pairing. This paper presents a ROS 2 frontier-allocation layer that combines a weighted frontier cost with global one-to-one matching. The study isolates global matching from sequential assignment while keeping the cost formulation and ROS 2 execution stack fixed. Three policies are evaluated on two indoor maps, four team sizes, and three seeds. Across 72 completed main-policy runs, global matching gives the lowest mean completion time, travelled distance, path overlap, and assignment conflict. Relative to sequential cost-based assignment, it reduces completion time by 24.3%, travelled distance by 14.2%, and path overlap by 65.8%, with lower final coverage under the same stopping rule. The results support a coordination-efficiency benefit in the tested ROS 2 simulations; broader claims require larger, heterogeneous, dynamic, and physical deployments.
Volume: 15
Issue: 3
Page: 589-596
Publish at: 2026-09-01

Graph-guided contrastive transformer architecture for robust and explainable network intrusion detection

10.11591/ijra.v15i3.pp698-708
Archana Jayapal , Kamalakkannan Somasundaram , Arun Kumar Ramamoorthy
Intrusion detection systems (IDS) are very instrumental in protecting contemporary network infrastructures against the ever-advancing cyberattacks. Conventional signature-based and machine learning-enabled IDS solutions frequently have difficulty when it comes to high false-positive rates, inability to flexibly adapt to novel attacks, and the lack of support for complex traffic dynamics. New deep learning architectures have better detection properties, yet are limited by feature overlap, temporality, and lack of extensiveness to generalization in changing network conditions. To overcome these issues, this paper presents a new graph-guided contrastive transformer-based intrusion detection system (GCT-IDS) which aims at improving detection accuracy and robustness and preserving real-time feasibility. The framework combines feature interaction by graph modeling, contrastive representation learning, and a sparse self-attention transformer to effectively learn global traffic relationships and behavioral variations. The CSE-CIC-IDS2018 data is used to test the proposed method in real network conditions.
Volume: 15
Issue: 3
Page: 698-708
Publish at: 2026-09-01

Human state digital twin architecture for physiology constrained adaptive robotic autonomy

10.11591/ijra.v15i3.pp503-518
Jaganathan Nirmaladevi , S. Saranya , Nidhi Mishra , Fazal Noorbasha , Manoharan Kavitha
Time-dependent human cognitive and physical conditions should be considered in adaptive autonomy in collaborative robotics to ensure safety, performance, and operator comfort. But most of the current methods depend on heuristic levels of workload or individual physiological measures and fail to combine human-state estimation with the control structure. This paper suggests a human state digital twin (HSDT) framework of physiological-constrained adaptive robotic autonomy. This framework is a combination of multimodal physiological measures of electrocardiography-based heart-rate variability, electromyography, electrodermal activity, and task context to predict latent human states such as cognitive workload, fatigue, and stress. These estimates are included in a twin-constrained adaptive autonomy controller, which varies the control authority of the robot in real time. The simulated collaborative manipulation situation is tested in the proposed framework in different workload and fatigue conditions. It has demonstrated better tracking performance, less torque ripple, reduced operator stress indicators and less torque ripple. The paper presents a controlled and reconfigurable physiology-aware adaptive autonomy control architecture and a robotics-based digital twin control design framework of human-aware control design.
Volume: 15
Issue: 3
Page: 503-518
Publish at: 2026-09-01

PIKER-NET: Multi-class retinal disease classification using Pied Kingfisher optimization-based improved residual network

10.11591/ijra.v15i3.pp678-689
Lissy Devasahayam , Ramya Devi Murugadasan , Anandhi Samuel Vijayalakshmi , Chanthiya Puhalenthi , Ramnath Muthusamy , Ahilan Appathurai , Natarajan Mohana Suganthi
Retinal diseases are vision-threatening conditions, including age-related macular degeneration (ARMD), diabetic retinopathy (DR), and glaucoma, that require early and accurate detection to prevent blindness. However, existing methods often struggle with limited feature representation, high inter-class similarity, intra-class variability, and reduced performance in handling noisy and low-quality retinal images. To address these challenges, a novel PIKER-NET framework is proposed for accurate multi-class retinal disease classification. The input fundus images from the RFMiD dataset are pre-processed using a scalable range adaptive bilateral (SCRAB) filter to enhance image clarity by preserving edges while reducing noise. The Improved Residual Network-Rescaled (ImResNet-RS) integrated with Temporal Attention is then employed to extract deep hierarchical features with enhanced discriminative power. Pied Kingfisher Optimization (PKO) algorithm is utilized for feature selection, effectively reducing redundant information while retaining the most relevant features. Residual Multilayer Perceptron (ResMLP) is used to classify retinal diseases into ARMD, branch retinal vein occlusion (BRVO), diabetic neuropathy (DN), DR, healthy, and myopia (MYA). The PIKER-NET achieves an overall accuracy of 98.14% and F1-score of 97.06%. The PIKER-NET approach improves overall accuracy by 3.24%, 4.24%, 6.23%, and 2.00% compared to EyeDeep-Net, IDL-MRDD, DeepDiabetic, and VisionDeep-AI, respectively. The proposed approach has strong clinical relevance by supporting earlier disease screening, reducing misdiagnosis, and enabling faster diagnosis to assist ophthalmologists in improving patient outcomes.
Volume: 15
Issue: 3
Page: 678-689
Publish at: 2026-09-01

DIETARY-GDM: AI-powered dual-source framework for personalized dietary recommendation in gestational diabetes management

10.11591/ijra.v15i3.pp709-719
Karthick Myilvahanan Jothivel , Arun Madhan , Sathishkumar Krishnaveni , Suganya Arumugam , Kamalakannan Subbiah , Dharmaraj Thalaihatty Belli
Gestational diabetes mellitus (GDM) is a condition of glucose intolerance that occurs during pregnancy, characterized by elevated blood glucose levels. However, the existing approach was validated on limited datasets, which may affect its generalizability to diverse real-world clinical settings, and its real-time deployment feasibility has not been thoroughly examined. In this research, a novel DIETARY-GDM (DIETARY food recommendation for GDM) has been proposed for managing diabetes in pregnant women through personalized diet food recommendations. Wavelet denoising and text cleaning are employed to remove noise in the signal and user input to enhance the quality. The enhanced signals and text are then extracted using the convolutional neural network-long short-term memory (CNN-LSTM) and universal sentence encoding (USE) to effectively analyze the body condition of the pregnant women. A deep neural network (DNN) integrated with a Bayesian optimization algorithm (BOA) is utilized to predict personalized dietary food recommendations. Finally, an artificial intelligence (AI)-BOT generates food suggestions for pregnant women to improve maternal health. From the experimental results, the DIETARY-GDM achieves an accuracy of 99.21%. The accuracy of the DIETARY-GDM was higher than that of the convolutional neural network (CNN), artificial neural network (ANN), gated recurrent unit (GRU), and long short-term memory (LSTM) by 7.98%, 5.65%, 3.35%, and 1.97%, respectively.
Volume: 15
Issue: 3
Page: 709-719
Publish at: 2026-09-01

Spatial-channel reconstruction for efficient multiscale attention in robotic object detection

10.11591/ijra.v15i3.pp544-552
Mohammed Maiza , Chahira Cherif , Samira Chouraqui , Abdelmalik Taleb-Ahmed
Real-time object detection is a core capability for autonomous robots, unmanned aerial vehicles (UAVs), and self-driving systems operating in resource-constrained environments. This paper presents spatial-channel enhanced multiscale attention (SCEMA), a novel lightweight attention module designed to enhance robotic perception while minimizing computational overhead for embedded deployment. SCEMA employs a parallel dual-branch architecture that synergistically combines spatial-channel reconstruction with multiscale attention mechanisms. When integrated into the YOLOv8n framework (3.01M baseline parameters), the proposed YOLO-SCEMA model achieves significant performance gains across multiple challenging benchmarks relevant to robotics automation. Experiments on an NVIDIA RTX 4080 GPU demonstrate that on the ExDark dataset, YOLO-SCEMA improves mAP@50 by 7.37% over the baseline (69.07% to 76.44%) while reducing parameters by 36.88% (3.01M to 1.90M) and computational cost by 8.64% (8.1 to 7.4 GFLOPs). Consistent improvements are also observed on VisDrone2019 (+3.24% mAP@50) and FYP (+1.50% mAP@50) datasets. Comparative analysis demonstrates that YOLOSCEMA achieves superior accuracy-efficiency trade-offs, making it particularly suitable for deployment in low-light conditions, dense scenes, and complex structural environments for autonomous navigation, robotic surveillance, and industrial automation applications.
Volume: 15
Issue: 3
Page: 544-552
Publish at: 2026-09-01
1 2 3
...
2091 2092
Show 23 of 2092

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration