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

31,456 Article Results

Mixed attention mechanism on ResNet-DeepLabV3+ for paddy field segmentation

10.12928/telkomnika.v23i6.26829
Alya; University of Indonesia Khairunnisa Rizkita , Masagus Muhammad; University of Indonesia Luthfi Ramadhan , Yohanes Fridolin; University of Indonesia Hestrio , Muhammad Hannan; University of Indonesia Hunafa , Danang Surya; National Research and Innovation Agency Candra , Wisnu; University of Indonesia Jatmiko
Rice cultivation monitoring is crucial for Indonesia, where paddy field areas de clined by 2.45% according to the Central Bureau of Statistics due to land func tion changes and shifting crop preferences. Regular monitoring of paddy field distribution is essential for understanding agricultural land utilization by farmers and landowners. Satellite imagery has become increasingly common for agricul tural land observation, but traditional neural networks alone provide insufficient segmentation accuracy. This study proposes an enhanced deep learning architec ture combining residual network (ResNet)-DeepLabV3+ with coordinate atten tion (CA) and spatial group-wise enhancement (SGE) modules. The attention mechanisms establish direct connections between context vectors and inputs, enabling the model to prioritize relevant spatial and spectral features for precise paddy field identification. The CA module enhances spectral feature discrim ination, whereas the SGE improves spatial characteristic representation. The experimental results demonstrate superior performance over the baseline meth ods, achieving intersection over union (IoU) of 0.85, dice coefficient of 0.89, and accuracy of 0.95. The proposed mixed attention mechanism significantly improves the accuracy and efficiency of automatic crop area identification from satellite imagery.
Volume: 23
Issue: 6
Page: 1611-1625
Publish at: 2025-12-01

Kannada handwritten numeral recognition through deep learning and optimized hyperparameter tuning

10.11591/ijai.v14.i6.pp5038-5048
Ujwala B. S. , Pramod Kumar S. , H. R. Mahadevaswamy , Sumathi K.
The classification of handwritten numerals is a vital and challenging task in developing automated systems, including postal address sorting and license plate recognition. The present study elucidates a new methodology for recognizing Kannada handwritten numerals using deep learning ResNet and VGG architecture with transfer learning. The challenge in Kannada handwritten recognition is complicated structural hierarchy and large vocabulary. The major problem in deep neural networks is vanishing gradient, which can lead to degradation in character recognition, and was addressed using our new methodology using ResNet architecture. We apply the proposed ResNet method in various real-world applications and compare it with convolutional neural networks (CNN) architecture, VGG. The experiment was implemented with the Google Colab software version on a self-created dataset, with handwritten Kannada numerals fed as the input to the recognition process. Our proposed method achieved a high accuracy of 99.20% on training samples and a generalization accuracy of 97.5% on test samples, indicating our method's effectiveness in recognizing handwritten Kannada numerals.
Volume: 14
Issue: 6
Page: 5038-5048
Publish at: 2025-12-01

Hybrid CNBLA architecture for accurate earthquake magnitude forecasting

10.11591/ijece.v15i6.pp5879-5893
Somia A. Shams , Asmaa Mohamed , Abeer S. Desuky , Gaber A. Elsharawy , Rania Salah El-Sayed
Earthquake prediction in seismology is challenging due to sudden events and lack of warnings, requiring rapid detection and accurate parameter estimation for real-time applications. This study proposed a novel automatic earthquake detection model to enhance the processing and analysis of seismic data. The hybrid model comprises convolutional layers, normalization techniques, bidirectional long short-term memory (Bi-LSTM) networks, and attention mechanisms, collectively referred to as the hybrid convolutional–normalization–BiLSTM–attention (CNBLA) model. The attention mechanism allows the model to focus on critical segments of seismic sequences, while layer normalization stabilizes training by normalizing activations, thus reducing the effects of input scale variations. This dual approach mitigates the impact of input scale variations and enhances the model’s ability to effectively decode complex temporal patterns. The hybrid CNBLA model optimizes the extraction and processing of temporal features from raw waveforms recorded at single stations, thereby improving the accuracy and efficiency of seismic magnitude estimation. The proposed model is evaluated using two datasets: the STEAD and USGS achieving a mean square error (MSE) values 0.054 and 0.0843 and a mean absolute error (MAE) 0.15 and 0.2526 respectively. The hybrid CNBLA model outperforms two baseline models and five state-of-the-art approaches in earthquake magnitude estimation, improving seismic monitoring and early warning systems.
Volume: 15
Issue: 6
Page: 5879-5893
Publish at: 2025-12-01

Performance enhancement of PV generator using a sensor based dual axis solar tracking system in Algeria

10.12928/telkomnika.v23i6.26872
Sakina; Udes/Centre De Développement des Énergies Renouvelables (CDER) Atoui , Harb; University of Algiers 1 Benyoucef Benkhedda Hadjer , Belaïd; University of Algiers 1 Benyoucef Benkhedda Abdelghani
This article presents the implementation of a two-axis solar tracking system and its impacts to increase the performance of the photovoltaic system in northern Algeria. The system enhances the efficiency of solar systems by optimizing their exposure to sunlight making the sunbeam perpendicular to solar panel. The main objective of the study is to develop a technically proficient and economically viable solution to increase solar energy production. The design relies on integrating light sensors and motors controlled by an Arduino board, enabling automatic adjustment of solar panel positions. This approach offers dynamic and precise orientation, based on light dependent resistor (LDR) sensor design and threshold value, resulting in a significant increase in energy output. The results show that the dual-axis solar tracking system can capture 60.64% more solar energy, taking into account the power consumption of the two electric actuators. The findings of this study will positively influence the promotion of clean and sustainable energy sources while providing a practical solution for more efficient utilization of solar energy in Algeria.
Volume: 23
Issue: 6
Page: 1706-1717
Publish at: 2025-12-01

Computationally efficient pixelwise deep learning architecture for accurate depth reconstruction for single-photon LiDAR

10.11591/ijece.v15i6.pp5934-5941
Yu Zhang , Yiming Zheng
This work introduces a compact deep learning architecture for depth image reconstruction from time-resolved single-photon histograms. Unlike most deep learning approaches that mainly rely on 3D convolutions, our network is implemented purely with 1D convolutions without assistance from other sensors or pre-processing. Both synthetic and real datasets were used to evaluate the accuracy of our model for challenging signal-to-background ratios (SBRs), ranging from 5:1 to 1:1. Conventional maximum likelihood (ML) and another photon-efficient optimization-based algorithm were adopted for performance comparisons. Results from synthetic data show that our model achieves lower mean absolute error (MAE). Additionally, results from real data indicate that our model exhibits better reconstruction for high-ambient effects and provides better spatial information. Unlike existing 3D deep learning models, we process pixel-wise histograms continuously, rather than splitting the point cloud and stitching them afterward, which saves memory and computational resources, thereby laying a foundation for real-world embedded applications.
Volume: 15
Issue: 6
Page: 5934-5941
Publish at: 2025-12-01

Performance analysis of a cascaded dual full bridges 5, 7, and 9 levels inverter: experimental validation

10.11591/ijpeds.v16.i4.pp2464-2475
Nabil Saidani , Rachid El Bachtiri , Abdelaziz FRI , Karima El Hammoumi
Cascaded full-bridge inverter is a suitable topology for grid-connected applications due to its ability to generate an output voltage waveform that closely resembles a sine wave, resulting in lower total harmonic distortion (THD) factors. This article proposes the use of the selective harmonic elimination (SHE) technique to produce a 5-level voltage using a symmetrical inverter and 7 and 9-level voltage using an asymmetrical inverter composed of only two full bridges loaded by an RL circuit of 51.4 Ω and 200 mH. The study primarily focuses on analysing the impact of the number of levels on the power quality of the inverter. This includes investigating the effects of the fundamental magnitude on the produced power, as well as measuring losses in the inverter, power factor, THD factor, and fundamental magnitude for each level configuration. The study demonstrates that asymmetrical MLIs lower THD (10.9% vs. 16.7%) and increasing voltage levels enhance waveform quality but slightly reduce the fundamental voltage magnitude, impacting AC power output. The simulation analysis has been conducted using the PSIM environment, and the results have been validated through experimental measurements.
Volume: 16
Issue: 4
Page: 2464-2475
Publish at: 2025-12-01

Optimization of water resource management in crops using satellite technology and artificial intelligence techniques

10.11591/ijece.v15i6.pp5847-5853
Erick Salvador Reyes-Galván , Fredy Alexander Bolivar-Gomez , Yeison Alberto Garcés-Gómez
This study aims to optimize water consumption in avocado crops through the application of satellite technology, machine learning algorithms, and precise climate data from the climate hazards group infrared precipitation with stations (CHIRPS) system. Crop classification in satellite images is conducted using the random forest algorithm, enabling detailed categorization of cultivated areas, urban land, soil, and vegetation, with a specific focus on avocados due to their high-water demand. Given its economic importance and status as one of the most water-intensive crops, avocado cultivation presents a critical challenge for agricultural sustainability. To validate predictive models and ensure classification accuracy, advanced evaluation methodologies such as the confusion matrix and Cohen's kappa index are utilized, quantifying the precision and reliability of the results. This estimation of water consumption under deficit and surplus conditions offers key insights for efficient water management in avocado cultivation. The results generated can enhance agricultural efficiency by aligning water use with the crop’s actual requirements, thereby contributing to the reduction of its water footprint.
Volume: 15
Issue: 6
Page: 5847-5853
Publish at: 2025-12-01

Designing, developing and analyzing of a rectangular-shaped patch antenna at 3.5 GHz for 5G applications at S band

10.11591/ijece.v15i6.pp5422-5432
Sukanto Halder , Md. Sohel Rana , Md Abdul Ahad , Md. Shehab Uddin Shahriar , Md. Abdulla Al Mamun , Md. Mominur Rahaman , Omer Faruk , Md. Eftiar Ahmed
This research study focuses on the design and analysis of two distinct patch antennas for 5G applications at 3.5 GHz. Rogers RT5880 served as the foundational material for antenna designs I and II. A 50 Ω feed line is utilized to supply both antennas. According to the calculations, Design I exhibits a reflection coefficient (S11) of -32.98 dB, a voltage standing wave ratio of 1.045, a gain of 7.81 dBi, an efficiency of 89.2%, and a surface current of 66.82 A/V. Design II has a reflection coefficient (S11) of 34.98 dB, voltage standing wave ratio (VSWR) of 1.036, gain of 8.78 dBi, efficiency of 89.87%, and surface current of 62.7 A/V. Among the two antenna designs, design II outperformed design I, and the results indicate that the antenna fulfilled the designated purpose. The novelties of the proposed paper are to design two different patch antennas using same materials and highlight the performance of the design parameters. Design II is proficient in supporting 5G services owing to its advantageous performance. In addition, S11 of the antenna is reduced to bring the VSWR value is close to 1. Also, improve gain, directivity and efficiency by bringing the antenna impedance matching close to 50 Ω.
Volume: 15
Issue: 6
Page: 5422-5432
Publish at: 2025-12-01

Enhanced ankle physiotherapy robot with electromyography - triggered ankle velocity control

10.11591/ijece.v15i6.pp5314-5326
Dimas Adiputra , Radithya Anjar Nismara , Muhammad Rafli Ramadhan Lubis , Nur Aliffah Rizkianingtyas , Kensora Bintang Panji Satrio , Rangga Roospratama Arif , Annisa Salsabila
Previous ankle physiotherapy robots, called picobot rely on predefined trajectories continuous passive movement without considering patient intent, limiting the encouragement of user-intent motion. This study then integrates electromyography (EMG) signals as triggers into picobot with an ankle velocity-based control system. The upgraded robot activates movement in specific gait phases based on muscle activity, synchronizing therapy with the patient’s intent. Functionality test on 7 young male healthy subjects investigates leg muscles, such as Tibialis Anterior, Soleus, and Gastrocnemius muscles for the most significantly contribute to ankle movements. Then, the muscle is tested to trigger picobot movements. Functionality tests revealed the Tibialis muscle significantly contributes to gait phases 2, the Soleus is prominent in phases 3 and 4, and gastrocnemius is active on phase 1. The robot successfully performs plantarflexion when EMG signals exceed a 1.58 V threshold, reaching a target position of -0.11 rad at a constant velocity of -0.62 rad/s. These findings establish a foundation for future trials since patient testing has not yet been conducted. By promoting active participation, this innovation has the potential to enhance rehabilitation outcomes. Incorporating user-intent triggers may accelerate recovery and improve healthcare accessibility in Indonesia, offering a significant advancement in physiotherapy technologies.
Volume: 15
Issue: 6
Page: 5314-5326
Publish at: 2025-12-01

Enhanced incremental conductance MPPT method for maximizing photovoltaic power generation

10.11591/ijpeds.v16.i4.pp2757-2767
Asnil Asnil , Refdinal Nazir , Krismadinata Krismadinata , Muhammad Nasir
This research proposes an enhanced maximum power point tracking (MPPT) algorithm that integrates the variable step size (VSS) method to significantly improve power extraction from photovoltaic (PV) systems. The primary objective is to optimize performance under dynamic environmental conditions. Through comprehensive experimental studies, the proposed algorithm’s performance was evaluated and directly compared against conventional incremental conductance (INC) and perturb and observe (P&O) algorithms. The results demonstrate a substantial increase in power generation, with the proposed algorithm delivering 18.79% more power compared to INC and 39.67% more power than P&O. These findings underscore the efficacy of the developed algorithm at improving the efficiency and robustness of PV power generation, particularly in variable operating environments.
Volume: 16
Issue: 4
Page: 2757-2767
Publish at: 2025-12-01

Time-domain performance of QBC with self-lift circuit

10.11591/ijpeds.v16.i4.pp2491-2499
Subbulakshmy Ramamurthi , Palani Velmurugan , Shobana Devendiren , Soundarapandiyan Manivannan
This study examines the performance of a high-gain quadratic boost converter (QBC) coupled with a self-lift circuit under two control methodologies: sliding mode control (SMC) and fractional-order proportional integral derivative (FOPID) control. The QBC topology is used because it can boost voltage significantly, which is especially useful for renewable energy applications. Simulation studies show that both controllers can control the output voltage of the converter, but the FOPID controller works better in dynamic situations. In particular, it makes settling happen faster, cuts down on overshoot, and lowers steady-state error compared to the SMC method. The overall results show that the FOPID controller is a good choice for improving stability and transient response. This makes it a good choice for advanced high-performance power electronic systems.
Volume: 16
Issue: 4
Page: 2491-2499
Publish at: 2025-12-01

Design and simulation of inductive power transfer system using hybrid compensation topologies

10.11591/ijpeds.v16.i4.pp2453-2463
Noor Sami , Harith Al-Badrani
This research addresses the principle of wireless power transfer (WPT). The system is primarily based on inductive power transfer (IPT). IPT is a recent technology that enables electrical power to be transferred between two coils via a magnetic field without the need for physical conductors. This method is particularly useful in applications where conventional wires cannot be used, such as biomedical implants, electric vehicles, and consumer electronics. Existing advances in system design, magnetic materials, and compensation topologies have significantly improved system performance and expanded their application range. Main challenges in IPT systems include improving efficiency and transmission distance. Hybrid compensation techniques in IPT systems have emerged as a promising solution to enhance system stability and power transfer efficiency under varying load conditions. IPT systems ensure highly efficient battery transfer and charging. This paper presents the design and simulation of a 3.7 kW IPT system employing hybrid compensation topologies specifically inductor–capacitor–capacitor/series (LCC/S) and LCC/LCC configurations to enhance power transfer efficiency and maintain zero phase angle (ZPA) operation. The proposed system is simulated using ANSYS Maxwell and MATLAB/Simulink to evaluate voltage gain, resonant behavior, and power output under varying load conditions. The LCC/LCC topology demonstrates superior load-independent ZPA characteristics and improved receiver-side voltage stability. Simulation results confirm that both configurations achieve high efficiency and robust power transfer over an air gap of 100 mm, with the LCC/LCC system showing better tolerance to misalignment. These findings suggest that hybrid compensation topologies are viable candidates for medium-power wireless charging systems in electric vehicles and industrial automation.
Volume: 16
Issue: 4
Page: 2453-2463
Publish at: 2025-12-01

Adaptive control strategies for enhancing the performance and stability of renewable energy systems

10.11591/ijpeds.v16.i4.pp2411-2418
Sreedevi Kunumalla , Durgam Rajababu , A. V. V. Sudhakar
The comparison done in this paper uses two control strategies for a solar-fed three-phase inverter, one with standard sinusoidal pulse width modulation (SPWM) and the other with unified space vector PWM (USPWM), equipped with adaptive voltage control (AVC) and a load current observer (LCO). A photovoltaic (PV) source feeds into a controlled DC link, which in turn supplies a DC voltage to the inverter powering an RL load. From the simulation, it is clear that although SPWM meets the output standards, it has increased total harmonic distortion (THD) and slower transient response. Alternatively, using USPWM control decreased THD to 3.01% and made the system respond in 11 ms, compared to 22 ms before. The new method provides better efficiency and better power quality when used with dynamic loads.
Volume: 16
Issue: 4
Page: 2411-2418
Publish at: 2025-12-01

Design of a static synchronous compensator for the north-south high-speed railway system

10.11591/ijpeds.v16.i4.pp2369-2380
An Thi Hoai Thu Anh , Tran Hung Cuong
The modern high-speed rail system plays a crucial role in driving the nation’s economic development. The problem of voltage imbalance caused by intermittent load movements is a significant challenge for energy management and distribution. When electric trains are connected to the three-phase grid, power quality degradation occurs, resulting in distortion and imbalance of the three-phase grid current and voltage, which in turn increases operating costs. This paper has proposed a linear control method using a PI controller for a static synchronous compensator (STATCOM) to directly control the amount of reactive power loss for electric trains. This solution will also bring good and stable voltage quality to electric trains so that electric trains can operate for a long time. The STATCOM device in this paper is a three-phase voltage source converter with a simple structure and can be easily controlled. This is considered a simple and effective solution to balance voltage, improve power factor, and enhance harmonic quality for railway trains, thereby achieving an optimal operating solution. This discussion can be simulated using MATLAB/Simulink software to determine the operation and control steps for STATCOM, thereby improving the quality of the power system. The simulation results of current, voltage, and reactive power response are presented. The simulation results have demonstrated that the proposed algorithm successfully achieves the set goals of ensuring voltage stability and providing the necessary amount of reactive power for the train, thereby improving the quality of the power grid for the North-South high-speed train in Vietnam.
Volume: 16
Issue: 4
Page: 2369-2380
Publish at: 2025-12-01

Machine learning model for accurate prediction of coronary artery disease by incorporating error reduction methodologies

10.11591/ijece.v15i6.pp5655-5666
Santhosh Gupta Dogiparthi , Jayanthi K. , Ajith Ananthakrishna Pillai , K. Nakkeeran
Coronary artery disease (CAD) remains a leading cause of mortality worldwide, with an especially high burden in developing countries such as India. In light of increasing patient loads and limited medical resources, there is an urgent need for accurate and reliable diagnostic support systems. This study introduces a machine learning (ML) framework that aims to enhance CAD prediction accuracy by specifically addressing the reduction of false negatives (FN), which are critical in medical diagnostics. Utilizing a stacked ensemble model comprising five base classifiers and a meta-classifier, the framework integrates cost-sensitive learning, classification threshold tuning, engineered features, and manual weighting strategies. The model was developed using a clinically acquired dataset from the Jawaharlal Institute of postgraduate medical education and research (JIPMER), consisting of 428 patient records with 36 original features. Evaluation metrics show that the proposed model achieved an accuracy of 92.19%, sensitivity of 98%, and an F1-score of 95.15%. These improvements are significant in a clinical context, potentially reducing missed diagnoses and improving patient outcomes. The model is intended for deployment in cardiology outpatient settings and demonstrates a scalable, adaptable approach to medical diagnostics.
Volume: 15
Issue: 6
Page: 5655-5666
Publish at: 2025-12-01
1 2 3
...
2097 2098
Show 164 of 2098

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