Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learni...Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learning(DL)approaches often face several limitations,including inefficient feature extraction,class imbalance,suboptimal classification performance,and limited interpretability,which collectively hinder their deployment in clinical settings.To address these challenges,we propose a novel DL framework for heart disease prediction that integrates a comprehensive preprocessing pipeline with an advanced classification architecture.The preprocessing stage involves label encoding and feature scaling.To address the issue of class imbalance inherent in the personal key indicators of the heart disease dataset,the localized random affine shadowsampling technique is employed,which enhances minority class representation while minimizing overfitting.At the core of the framework lies the Deep Residual Network(DeepResNet),which employs hierarchical residual transformations to facilitate efficient feature extraction and capture complex,non-linear relationships in the data.Experimental results demonstrate that the proposed model significantly outperforms existing techniques,achieving improvements of 3.26%in accuracy,3.16%in area under the receiver operating characteristics,1.09%in recall,and 1.07%in F1-score.Furthermore,robustness is validated using 10-fold crossvalidation,confirming the model’s generalizability across diverse data distributions.Moreover,model interpretability is ensured through the integration of Shapley additive explanations and local interpretable model-agnostic explanations,offering valuable insights into the contribution of individual features to model predictions.Overall,the proposed DL framework presents a robust,interpretable,and clinically applicable solution for heart disease prediction.展开更多
Nuclei segmentation is a challenging task in histopathology images.It is challenging due to the small size of objects,low contrast,touching boundaries,and complex structure of nuclei.Their segmentation and counting pl...Nuclei segmentation is a challenging task in histopathology images.It is challenging due to the small size of objects,low contrast,touching boundaries,and complex structure of nuclei.Their segmentation and counting play an important role in cancer identification and its grading.In this study,WaveSeg-UNet,a lightweight model,is introduced to segment cancerous nuclei having touching boundaries.Residual blocks are used for feature extraction.Only one feature extractor block is used in each level of the encoder and decoder.Normally,images degrade quality and lose important information during down-sampling.To overcome this loss,discrete wavelet transform(DWT)alongside maxpooling is used in the down-sampling process.Inverse DWT is used to regenerate original images during up-sampling.In the bottleneck of the proposed model,atrous spatial channel pyramid pooling(ASCPP)is used to extract effective high-level features.The ASCPP is the modified pyramid pooling having atrous layers to increase the area of the receptive field.Spatial and channel-based attention are used to focus on the location and class of the identified objects.Finally,watershed transform is used as a post processing technique to identify and refine touching boundaries of nuclei.Nuclei are identified and counted to facilitate pathologists.The same domain of transfer learning is used to retrain the model for domain adaptability.Results of the proposed model are compared with state-of-the-art models,and it outperformed the existing studies.展开更多
Roll coating is a vital industrial process used in printing,packaging,and polymer film production,where maintaining a uniform coating is critical for product quality and efficiency.This work models non-isothermal Carr...Roll coating is a vital industrial process used in printing,packaging,and polymer film production,where maintaining a uniform coating is critical for product quality and efficiency.This work models non-isothermal Carreau fluid flow between a rotating roll and a stationary wall under fixed boundary constraints to evaluate how non-Newtonian and thermal effects influence coating performance.The governing equations are transformed into non-dimensional form and simplified using lubrication approximation theory.Approximate analytical solutions are obtained via the perturbation technique,while numerical results are computed using both the finite difference method and the BVPMidrich technique.Furthermore,Response surface methodology(RSM)is employed for optimization and sensitivity analysis.Analytical and numerical results show strong agreement(<1%deviation).The model predicts coating thickness 0.55≤λ≤0.64,power input 1.05≤P_(w)≤1.99,and separation force 0.91≤S_(f)≤1.82 for 0.1≤We≤0.9 and 0.01≤F≤0.09.Increasing We enhances the coating thickness and power input but reduces velocity and separation force.The findings provide physical insight into elastic and viscous effects in roll coating,providing insight for optimizing coating uniformity,minimizing wear,improving industrial coating processes,and extending substrate lifespan.展开更多
Arterial stenosis is a critical condition with increasing prevalence among pediatric patients and young adults,making its investigation highly significant.Despite extensive studies on blood flow dynamics,limited resea...Arterial stenosis is a critical condition with increasing prevalence among pediatric patients and young adults,making its investigation highly significant.Despite extensive studies on blood flow dynamics,limited research addresses the combined effects of nanoparticles and arterial curvature on unsteady pulsatile flow through multiple stenoses.This study aims to analyze the influence of nanoparticles on blood flow characteristics in realistic curved arteries with mild to severe overlapped constrictions.Using curvilinear coordinates,the thermal energy and momentum equations for nanoparticle-laden blood were derived,and numerical results were obtained through an explicit finite difference method.Key findings reveal that nanoparticle injections reduce blood temperature intensity,while arterial curvature strongly affects flow symmetry.Moreover,temperature,axial velocity,wall shear stress,and volumetric flow rate decrease significantly in severe stenosis compared to mild and moderate cases.These results provide new insights into nanoparticle-assisted blood flow under complex stenotic conditions and may contribute to improved diagnostic and therapeutic strategies for cardiovascular diseases.展开更多
The increased accessibility of social networking services(SNSs)has facilitated communication and information sharing among users.However,it has also heightened concerns about digital safety,particularly for children a...The increased accessibility of social networking services(SNSs)has facilitated communication and information sharing among users.However,it has also heightened concerns about digital safety,particularly for children and adolescents who are increasingly exposed to online grooming crimes.Early and accurate identification of grooming conversations is crucial in preventing long-term harm to victims.However,research on grooming detection in South Korea remains limited,as existing models trained primarily on English text and fail to reflect the unique linguistic features of SNS conversations,leading to inaccurate classifications.To address these issues,this study proposes a novel framework that integrates optical character recognition(OCR)technology with KcELECTRA,a deep learning-based natural language processing(NLP)model that shows excellent performance in processing the colloquial Korean language.In the proposed framework,the KcELECTRA model is fine-tuned by an extensive dataset,including Korean social media conversations,Korean ethical verification data from AI-Hub,and Korean hate speech data from Hug-gingFace,to enable more accurate classification of text extracted from social media conversation images.Experimental results show that the proposed framework achieves an accuracy of 0.953,outperforming existing transformer-based models.Furthermore,OCR technology shows high accuracy in extracting text from images,demonstrating that the proposed framework is effective for online grooming detection.The proposed framework is expected to contribute to the more accurate detection of grooming text and the prevention of grooming-related crimes.展开更多
In this article,a well-known anisotropic solution,the Tolman-Finch-Skea(TFS)solution,is studied using the gravitational decoupling approach within the framework of 4D Einstein-Gauss-Bonnet(EGB)gravity.The radial metri...In this article,a well-known anisotropic solution,the Tolman-Finch-Skea(TFS)solution,is studied using the gravitational decoupling approach within the framework of 4D Einstein-Gauss-Bonnet(EGB)gravity.The radial metric potential is modified linearly through the minimal geometric deformation approach,while the temporal component of the metric remains unchanged.The system of EGB field equations is decomposed into two distinct sets of field equations:one corresponding to the standard energy-momentum tensor and the other associated with an external gravitational source.The first system is solved using the aforementioned known solution,while the second is closed by imposing the mimic constraint on pressure.Moreover,the junction conditions at the inner and outer surfaces of the stellar object are examined,considering the Boulware-Deser 4D space-time as the external geometry.The physical properties of the stellar model are analyzed using parameters such as energy conditions,causality conditions,compactness,and redshift.展开更多
Neodymium chromium oxide(NdCrO_(3))and NdCrO_(3)/graphene oxide(GO)nanocomposite were synthesized via sol-gel and co-precipitation techniques for being used in high-perfo rmance supercapacitors and for the possible ap...Neodymium chromium oxide(NdCrO_(3))and NdCrO_(3)/graphene oxide(GO)nanocomposite were synthesized via sol-gel and co-precipitation techniques for being used in high-perfo rmance supercapacitors and for the possible application in ultraviolet(UV)materials.Herein the systematic synthesis approach was adopted,which enhances the optical and electrical properties of the grown wide band-gap composite nanomaterial.Structural characterization of the grown materials was attempted using X-ray diffraction(XRD)and scanning electron microscopy(SEM).Most importantly the electrochemical analysis of the grown samples was carried out by employing a glassy carbon electrode and 3 mol/L KOH electrolyte,which demonstrates significant improvements in a specific capacitance of approximately360 F/g,an energy density of approximately 18 Wh/kg,and a maximum power density of approximately 257 W/kg,respectively.Moreover,NdCrO_(3)/GO nanocomposite maintains a cyclic stability of 97.6%after4000 cycles.Photoluminescence(PL)spectroscopy confirms the wide bandgap nature of the NdCrO_(3)and NdCrO_(3)/GO nanocomposite,indicating its potential application in UVC devices.These findings emphasize the potential of the NdCrO_(3)/GO nanocomposite in advancing efficient energy storage solutions and the possibility of being used in UVC technology.展开更多
Accurate estimation of the Direction-of-Arrival(DoA)of incident plane waves is essential for modern wireless communication,radar,sonar,and localization systems.Precise DoA information enables adaptive beamforming,spat...Accurate estimation of the Direction-of-Arrival(DoA)of incident plane waves is essential for modern wireless communication,radar,sonar,and localization systems.Precise DoA information enables adaptive beamforming,spatial filtering,and interference mitigation by steering antenna array beams toward desired sources while suppressing unwanted signals.Traditional one-dimensional Uniform Linear Arrays(ULAs)are limited to elevation angle estimation due to geometric constraints,typically within the range[0,π].To capture full spatial characteristics in environments with multipath and angular spread,joint estimation of both elevation and azimuth angles becomes necessary.However,existing 2D and 3D array geometries often entail increased hardware complexity and computational cost.This work proposes a novel and efficient framework for joint elevation and azimuth angle estimation using three spatially separated,parallel ULAs.The array configuration exploits spatial diversity and orthogonal projections to capture complete directional information with minimal structural overhead.A customized objective function based on the mean square error between measured and reconstructed array outputs is formulated to guide the estimation process.To solve the resulting non-convex optimization problem,three strategies are investigated:a global Genetic Algorithm(GA),a local Pattern Search(PS),and a hybrid GA-PS method that combines global exploration with local refinement.The proposed framework supports automatic pairing of elevation and azimuth angles,eliminating the need for manual post-processing.Extensive simulations validate the robustness,convergence,and accuracy of all three methods under varying signal-to-noise ratio conditions.Results confirm that the hybrid GA-PS approach achieves superior estimation performance and reduced computational complexity,making it well-suited for real-time and resource-constrained applications in next-generation sensing and communication systems.展开更多
The increasing demand for sustainable construction practices has led to growing interest in recycled aggregate concrete(RAC)as an eco-friendly alternative to conventional concrete.However,predicting its compressive st...The increasing demand for sustainable construction practices has led to growing interest in recycled aggregate concrete(RAC)as an eco-friendly alternative to conventional concrete.However,predicting its compressive strength remains a challenge due to the variability in recycled materials and mix design parameters.This study presents a robust machine learning framework for predicting the compressive strength of recycled aggregate concrete using feedforward neural networks(FFNN),Random Forest(RF),and XGBoost.A literature-derived dataset of 502 samples was enriched via interpolation-based data augmentation and modeled using five distinct optimization techniques within MATLAB’s Neural Net Fitting module:Bayesian Regularization,Levenberg-Marquardt,and three conjugate gradient variants—Powell/Beale Restarts,Fletcher-Powell,and Polak-Ribiere.Hyperparameter tuning,dropout regularization,and early stopping were employed to enhance generalization.Comparative analysis revealed that FFNN outperformed RF and XGBoost,achieving an R2 of 0.9669.To ensure interpretability,accumulated local effects(ALE)along with partial dependence plots(PDP)were utilized.This revealed trends consistent with the pre-existent domain knowledge.This allows estimation of strength from the properties of the mix without extensive lab testing,permitting designers to track the performance and sustainability trends in concrete mix designs while promoting responsible construction and demolition waste utilization.展开更多
Fog computing is a key enabling technology of 6G systems as it provides quick and reliable computing,and data storage services which are required for several 6G applications.Artificial Intelligence(AI)algorithms will ...Fog computing is a key enabling technology of 6G systems as it provides quick and reliable computing,and data storage services which are required for several 6G applications.Artificial Intelligence(AI)algorithms will be an integral part of 6G systems and efficient task offloading techniques using fog computing will improve their performance and reliability.In this paper,the focus is on the scenario of Partial Offloading of a Task to Multiple Helpers(POMH)in which larger tasks are divided into smaller subtasks and processed in parallel,hence expediting task completion.However,using POMH presents challenges such as breaking tasks into subtasks and scaling these subtasks based on many interdependent factors to ensure that all subtasks of a task finish simultaneously,preventing resource wastage.Additionally,applying matching theory to POMH scenarios results in dynamic preference profiles of helping devices due to changing subtask sizes,resulting in a difficult-to-solve,externalities problem.This paper introduces a novel many-to-one matching-based algorithm,designed to address the externalities problem and optimize resource allocation within POMH scenarios.Additionally,we propose a new time-efficient preference profiling technique that further enhances time optimization in POMH scenarios.The performance of the proposed technique is thoroughly evaluated in comparison to alternate baseline schemes,revealing many advantages of the proposed approach.The simulation findings indisputably show that the proposed matching-based offloading technique outperforms existing methodologies in the literature,yielding a remarkable 52 reduction in task latency,particularly under high workloads.展开更多
Due to the features and wide range of potential applications,cellulose ionogels are the subject of extensive research.Green celluloses have been employed as a three-dimensional skeleton network to restrict the ionic l...Due to the features and wide range of potential applications,cellulose ionogels are the subject of extensive research.Green celluloses have been employed as a three-dimensional skeleton network to restrict the ionic liquids(ILs)toward advanced ion-conductive ionogels.Diversiform cellulose ionogels with desirable perfor-mances,via physical/chemical reactions between cellulose and ILs,have been harvested,which have the po-tential to emerge as a bright star in the field of flexible electronics,such as sensors,electrolyte materials as power sources,and thermoelectric devices.Herein,a review regarding cellulose ionogels in terms of fundamental types of cellulose,formation strategies and mechanism,and principal properties is presented.Next,the diverse application prospects of cellulose ionogels in flexible electronics have been summarized.More importantly,the future challenges and advancing directions to be explored for cellulose ionogels are discussed.展开更多
This study explores the magnetohydrodynamic(MHD)boundary layer flow of a water-based Boger nanofluid over a stretching sheet,with particular focus on the influences of nanoparticle diameter,nanolayer effects,and therm...This study explores the magnetohydrodynamic(MHD)boundary layer flow of a water-based Boger nanofluid over a stretching sheet,with particular focus on the influences of nanoparticle diameter,nanolayer effects,and thermal radiation.The primary aim is to examine how variations in nanoparticle size and nanolayer thickness affect the hydrothermal behavior of the nanofluid.The model also incorporates the contributions of viscous dissipation and Joule heating within the heat transfer equation.The governing momentum and energy equations are converted into dimensionless partial differential equations(PDEs)using appropriate similarity variables and are numerically solved using the finite element method(FEM)implemented in MATLAB.Extensive validation of this method confirms its reliability and accuracy in numerical solutions.The findings reveal that increasing the diameter of copper nanoparticles significantly enhances the velocity profile,with a more pronounced effect observed at wider inter-particle spacings.A higher solvent volume fraction leads to decreased velocity and temperature distributions,while a greater relaxation time ratio improves velocity and temperature profiles due to the increased elastic response of the fluid.Moreover,enhancements in the magnetic parameter,thermal radiation,and Eckert number lead to an elevation in temperature profiles.Furthermore,higher nanolayer thickness reduces the temperature profile,whereas particle radius yields the opposite outcome.展开更多
The Langrial iron ore deposits,located in the villages of Dubran and Darkot in Hazara,Pakistan,were evaluated using remote sensing,magnetic,and geochemical investigations.Data from ASTER,Landsat-8,and Sentinel-2 satel...The Langrial iron ore deposits,located in the villages of Dubran and Darkot in Hazara,Pakistan,were evaluated using remote sensing,magnetic,and geochemical investigations.Data from ASTER,Landsat-8,and Sentinel-2 satellites were utilized and processed through techniques such as band ratio analysis,band compositing,and NDVI masking to reduce vegetation effects and to delineate various lithological units and mineralogical signatures within the study area.Magnetic anomalies revealed multiple levels of iron mineralization,with the hematite zone showing the most significant potential for high-grade iron ore.Geochemical analyses confirmed the presence of iron,along with minerals such as chromium,calcium,magnesium,and lead.In Dubran,mean iron concentrations are recorded at 370.94 mg/kg,whereas in Darkot,they reach up to 2052 mg/kg.The integration of remote sensing,magnetic,and geochemical data delineates key mineralized zones in the various parts of the study area.This research highlights the importance of combining geophysical and geochemical methodologies to refine mineral exploration efforts.The findings enhance our understanding of the Langrial iron ore deposits and highlight their economic potential for sustainable mining practices.This study will contribute to meeting the growing demand for iron ore resources and reducing Pakistan's reliance on imports,thereby promoting the sustainable development of local industries.展开更多
Smart learning environments have been considered as vital sources and essential needs in modern digital education systems.With the rapid proliferation of smart and assistive technologies,smart learning processes have ...Smart learning environments have been considered as vital sources and essential needs in modern digital education systems.With the rapid proliferation of smart and assistive technologies,smart learning processes have become quite convenient,comfortable,and financially affordable.This shift has led to the emergence of pervasive computing environments,where user’s intelligent behavior is supported by smart gadgets;however,it is becoming more challenging due to inconsistent behavior of Artificial intelligence(AI)assistive technologies in terms of networking issues,slow user responses to technologies and limited computational resources.This paper presents a context-aware predictive reasoning based formalism for smart learning environments that facilitates students in managing their academic as well as extra-curricular activities autonomously with limited human intervention.This system consists of a three-tier architecture including the acquisition of the contextualized information from the environment autonomously,modeling the system using Web Ontology Rule Language(OWL 2 RL)and Semantic Web Rule Language(SWRL),and perform reasoning to infer the desired goals whenever and wherever needed.For contextual reasoning,we develop a non-monotonic reasoning based formalism to reason with contextual information using rule-based reasoning.The focus is on distributed problem solving,where context-aware agents exchange information using rule-based reasoning and specify constraints to accomplish desired goals.To formally model-check and simulate the system behavior,we model the case study of a smart learning environment in the UPPAAL model checker and verify the desired properties in the model,such as safety,liveness and robust properties to reflect the overall correctness behavior of the system with achieving the minimum analysis time of 0.002 s and 34,712 KB memory utilization.展开更多
Vehicular networks enable seamless connectivity for exchanging emergency and infotainment content.However,retrieving infotainment data from remote servers often introduces high delays,degrading the Quality of Service(...Vehicular networks enable seamless connectivity for exchanging emergency and infotainment content.However,retrieving infotainment data from remote servers often introduces high delays,degrading the Quality of Service(QoS).To overcome this,caching frequently requested content at fog-enabled Road Side Units(RSUs)reduces communication latency.Yet,the limited caching capacity of RSUs makes it impractical to store all contents with varying sizes and popularity.This research proposes an efficient content caching algorithm that adapts to dynamic vehicular demands on highways to maximize request satisfaction.The scheme is evaluated against Intelligent Content Caching(ICC)and Random Caching(RC).The obtained results show that our proposed scheme entertains more contentrequesting vehicles as compared to ICC and RC,with 33%and 41%more downloaded data in 28%and 35%less amount of time from ICC and RC schemes,respectively.展开更多
An expanding human population and technological progress demand clean and effective energy-storing systems.Within the realm of energy-storing devices,supercapacitors(SCs)have grabbed huge focus owing to their high-pow...An expanding human population and technological progress demand clean and effective energy-storing systems.Within the realm of energy-storing devices,supercapacitors(SCs)have grabbed huge focus owing to their high-power density,unique cycling stability,and fast charging discharging capabilities.Electrode material has a prominent impact on the effectiveness of SCs.Several types of electrode materials have been used,encompassing varied metal oxides,activated carbon,conducting polymers,and MOFs.Metal organic frameworks(MOFs)are considered emerging electrode candidates,which could be ascribed to the tunable porosity,large surface areas,and designed morphology.This review shows a detailed analysis of various mono-,bi-,and tri-metallic MOFs along with derivatives in SC applications,their structural characteristics,and synthetic strategies.It also critically evaluates MOFs potential to boost the SC's energy density,power density,stability,and conductivity.Also,it underscores their significance in the establishment of future-oriented energy storage applications.展开更多
Diagnosing gastrointestinal tract diseases is a critical task requiring accurate and efficient methodologies.While deep learning models have significantly advanced medical image analysis,challenges such as imbalanced ...Diagnosing gastrointestinal tract diseases is a critical task requiring accurate and efficient methodologies.While deep learning models have significantly advanced medical image analysis,challenges such as imbalanced datasets and redundant features persist.This study proposes a novel framework that customizes two deep learning models,NasNetMobile and ResNet50,by incorporating bottleneck architectures,named as NasNeck and ResNeck,to enhance feature extraction.The feature vectors are fused into a combined vector,which is further optimized using an improved Whale Optimization Algorithm to minimize redundancy and improve discriminative power.The optimized feature vector is then classified using artificial neural network classifiers,effectively addressing the limitations of traditional methods.Data augmentation techniques are employed to tackle class imbalance,improving model learning and generalization.The proposed framework was evaluated on two publicly available datasets:Hyper-Kvasir and Kvasir v2.The Hyper-Kvasir dataset,comprising 23 gastrointestinal disease classes,yielded an impressive 96.0%accuracy.On the Kvasir v2 dataset,which contains 8 distinct classes,the framework achieved a remarkable 98.9%accuracy,further demonstrating its robustness and superior classification performance across different gastrointestinal datasets.The results demonstrate the effectiveness of customizing deep models with bottleneck architectures,feature fusion,and optimization techniques in enhancing classification accuracy while reducing computational complexity.展开更多
The complex morphological,anatomical,physiological,and chemical mechanisms within the aging brain have been the hot topic of research for centuries.The aging process alters the brain structure that affects functions a...The complex morphological,anatomical,physiological,and chemical mechanisms within the aging brain have been the hot topic of research for centuries.The aging process alters the brain structure that affects functions and cognitions,but the worsening of such processes contributes to the pathogenesis of neurodegenerative disorders,such as Alzheimer's disease.Beyond these observable,mild morphological shifts,significant functional modifications in neurotransmission and neuronal activity critically influence the aging brain.Understanding these changes is important for maintaining cognitive health,especially given the increasing prevalence of age-related conditions that affect cognition.This review aims to explore the age-induced changes in brain plasticity and molecular processes,differentiating normal aging from the pathogenesis of Alzheimer's disease,thereby providing insights into predicting the risk of dementia,particularly Alzheimer's disease.展开更多
In recent years,there has been a concerted effort to improve anomaly detection tech-niques,particularly in the context of high-dimensional,distributed clinical data.Analysing patient data within clinical settings reve...In recent years,there has been a concerted effort to improve anomaly detection tech-niques,particularly in the context of high-dimensional,distributed clinical data.Analysing patient data within clinical settings reveals a pronounced focus on refining diagnostic accuracy,personalising treatment plans,and optimising resource allocation to enhance clinical outcomes.Nonetheless,this domain faces unique challenges,such as irregular data collection,inconsistent data quality,and patient-specific structural variations.This paper proposed a novel hybrid approach that integrates heuristic and stochastic methods for anomaly detection in patient clinical data to address these challenges.The strategy combines HPO-based optimal Density-Based Spatial Clustering of Applications with Noise for clustering patient exercise data,facilitating efficient anomaly identification.Subsequently,a stochastic method based on the Interquartile Range filters unreliable data points,ensuring that medical tools and professionals receive only the most pertinent and accurate information.The primary objective of this study is to equip healthcare pro-fessionals and researchers with a robust tool for managing extensive,high-dimensional clinical datasets,enabling effective isolation and removal of aberrant data points.Furthermore,a sophisticated regression model has been developed using Automated Machine Learning(AutoML)to assess the impact of the ensemble abnormal pattern detection approach.Various statistical error estimation techniques validate the efficacy of the hybrid approach alongside AutoML.Experimental results show that implementing this innovative hybrid model on patient rehabilitation data leads to a notable enhance-ment in AutoML performance,with an average improvement of 0.041 in the R2 score,surpassing the effectiveness of traditional regression models.展开更多
Surveillance systems can take various forms,but gait-based surveillance is emerging as a powerful approach due to its ability to identify individuals without requiring their cooperation.In the existing studies,several...Surveillance systems can take various forms,but gait-based surveillance is emerging as a powerful approach due to its ability to identify individuals without requiring their cooperation.In the existing studies,several approaches have been suggested for gait recognition;nevertheless,the performance of existing systems is often degraded in real-world conditions due to covariate factors such as occlusions,clothing changes,walking speed,and varying camera viewpoints.Furthermore,most existing research focuses on single-person gait recognition;however,counting,tracking,detecting,and recognizing individuals in dual-subject settings with occlusions remains a challenging task.Therefore,this research proposed a variant of an automated gait model for occluded dual-subject walk scenarios.More precisely,in the proposed method,we have designed a deep learning(DL)-based dual-subject gait model(DSG)involving three modules.The first module handles silhouette segmentation,localization,and counting(SLC)using Mask-RCNN with MobileNetV2.The next stage uses a Convolutional block attention module(CBAM)-based Siamese network for frame-level tracking with a modified gallery setting.Following the last,gait recognition based on regionbased deep learning is proposed for dual-subject gait recognition.The proposed method,tested on Shri Mata Vaishno Devi University(SMVDU)-Multi-Gait and Single-Gait datasets,shows strong performance with 94.00%segmentation,58.36%tracking,and 63.04%gait recognition accuracy in dual-subject walk scenarios.展开更多
基金funded by Ongoing Research Funding Program for Project number(ORF-2025-648),King Saud University,Riyadh,Saudi Arabia.
文摘Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learning(DL)approaches often face several limitations,including inefficient feature extraction,class imbalance,suboptimal classification performance,and limited interpretability,which collectively hinder their deployment in clinical settings.To address these challenges,we propose a novel DL framework for heart disease prediction that integrates a comprehensive preprocessing pipeline with an advanced classification architecture.The preprocessing stage involves label encoding and feature scaling.To address the issue of class imbalance inherent in the personal key indicators of the heart disease dataset,the localized random affine shadowsampling technique is employed,which enhances minority class representation while minimizing overfitting.At the core of the framework lies the Deep Residual Network(DeepResNet),which employs hierarchical residual transformations to facilitate efficient feature extraction and capture complex,non-linear relationships in the data.Experimental results demonstrate that the proposed model significantly outperforms existing techniques,achieving improvements of 3.26%in accuracy,3.16%in area under the receiver operating characteristics,1.09%in recall,and 1.07%in F1-score.Furthermore,robustness is validated using 10-fold crossvalidation,confirming the model’s generalizability across diverse data distributions.Moreover,model interpretability is ensured through the integration of Shapley additive explanations and local interpretable model-agnostic explanations,offering valuable insights into the contribution of individual features to model predictions.Overall,the proposed DL framework presents a robust,interpretable,and clinically applicable solution for heart disease prediction.
文摘Nuclei segmentation is a challenging task in histopathology images.It is challenging due to the small size of objects,low contrast,touching boundaries,and complex structure of nuclei.Their segmentation and counting play an important role in cancer identification and its grading.In this study,WaveSeg-UNet,a lightweight model,is introduced to segment cancerous nuclei having touching boundaries.Residual blocks are used for feature extraction.Only one feature extractor block is used in each level of the encoder and decoder.Normally,images degrade quality and lose important information during down-sampling.To overcome this loss,discrete wavelet transform(DWT)alongside maxpooling is used in the down-sampling process.Inverse DWT is used to regenerate original images during up-sampling.In the bottleneck of the proposed model,atrous spatial channel pyramid pooling(ASCPP)is used to extract effective high-level features.The ASCPP is the modified pyramid pooling having atrous layers to increase the area of the receptive field.Spatial and channel-based attention are used to focus on the location and class of the identified objects.Finally,watershed transform is used as a post processing technique to identify and refine touching boundaries of nuclei.Nuclei are identified and counted to facilitate pathologists.The same domain of transfer learning is used to retrain the model for domain adaptability.Results of the proposed model are compared with state-of-the-art models,and it outperformed the existing studies.
基金supported by the Talent Project of Tianchi Young-Doctoral Program in Xinjiang Uygur Autonomous Region of China(No.51052401510)Natural Science Foundation General Project(Grant Number 2025D01C36)of the Xinjiang Uyghur Autonomous Region of China+1 种基金This study received financial support from the National Natural Science Foundation of Xinjiang Province(Grant Nos.2022TSYCTD0019 and 2022D01D32)the China Scholarship Council(CSC)(Grant No.2021SLJ009915).
文摘Roll coating is a vital industrial process used in printing,packaging,and polymer film production,where maintaining a uniform coating is critical for product quality and efficiency.This work models non-isothermal Carreau fluid flow between a rotating roll and a stationary wall under fixed boundary constraints to evaluate how non-Newtonian and thermal effects influence coating performance.The governing equations are transformed into non-dimensional form and simplified using lubrication approximation theory.Approximate analytical solutions are obtained via the perturbation technique,while numerical results are computed using both the finite difference method and the BVPMidrich technique.Furthermore,Response surface methodology(RSM)is employed for optimization and sensitivity analysis.Analytical and numerical results show strong agreement(<1%deviation).The model predicts coating thickness 0.55≤λ≤0.64,power input 1.05≤P_(w)≤1.99,and separation force 0.91≤S_(f)≤1.82 for 0.1≤We≤0.9 and 0.01≤F≤0.09.Increasing We enhances the coating thickness and power input but reduces velocity and separation force.The findings provide physical insight into elastic and viscous effects in roll coating,providing insight for optimizing coating uniformity,minimizing wear,improving industrial coating processes,and extending substrate lifespan.
文摘Arterial stenosis is a critical condition with increasing prevalence among pediatric patients and young adults,making its investigation highly significant.Despite extensive studies on blood flow dynamics,limited research addresses the combined effects of nanoparticles and arterial curvature on unsteady pulsatile flow through multiple stenoses.This study aims to analyze the influence of nanoparticles on blood flow characteristics in realistic curved arteries with mild to severe overlapped constrictions.Using curvilinear coordinates,the thermal energy and momentum equations for nanoparticle-laden blood were derived,and numerical results were obtained through an explicit finite difference method.Key findings reveal that nanoparticle injections reduce blood temperature intensity,while arterial curvature strongly affects flow symmetry.Moreover,temperature,axial velocity,wall shear stress,and volumetric flow rate decrease significantly in severe stenosis compared to mild and moderate cases.These results provide new insights into nanoparticle-assisted blood flow under complex stenotic conditions and may contribute to improved diagnostic and therapeutic strategies for cardiovascular diseases.
基金supported by the IITP(Institute of Information&Communications Technology Planning&Evaluation)-ITRC(Information Technology Research Center)grant funded by the Korean government(Ministry of Science and ICT)(IITP-2025-RS-2024-00438056).
文摘The increased accessibility of social networking services(SNSs)has facilitated communication and information sharing among users.However,it has also heightened concerns about digital safety,particularly for children and adolescents who are increasingly exposed to online grooming crimes.Early and accurate identification of grooming conversations is crucial in preventing long-term harm to victims.However,research on grooming detection in South Korea remains limited,as existing models trained primarily on English text and fail to reflect the unique linguistic features of SNS conversations,leading to inaccurate classifications.To address these issues,this study proposes a novel framework that integrates optical character recognition(OCR)technology with KcELECTRA,a deep learning-based natural language processing(NLP)model that shows excellent performance in processing the colloquial Korean language.In the proposed framework,the KcELECTRA model is fine-tuned by an extensive dataset,including Korean social media conversations,Korean ethical verification data from AI-Hub,and Korean hate speech data from Hug-gingFace,to enable more accurate classification of text extracted from social media conversation images.Experimental results show that the proposed framework achieves an accuracy of 0.953,outperforming existing transformer-based models.Furthermore,OCR technology shows high accuracy in extracting text from images,demonstrating that the proposed framework is effective for online grooming detection.The proposed framework is expected to contribute to the more accurate detection of grooming text and the prevention of grooming-related crimes.
基金partially supported by the National Natural Science Foundation of China under Grant No.11988101。
文摘In this article,a well-known anisotropic solution,the Tolman-Finch-Skea(TFS)solution,is studied using the gravitational decoupling approach within the framework of 4D Einstein-Gauss-Bonnet(EGB)gravity.The radial metric potential is modified linearly through the minimal geometric deformation approach,while the temporal component of the metric remains unchanged.The system of EGB field equations is decomposed into two distinct sets of field equations:one corresponding to the standard energy-momentum tensor and the other associated with an external gravitational source.The first system is solved using the aforementioned known solution,while the second is closed by imposing the mimic constraint on pressure.Moreover,the junction conditions at the inner and outer surfaces of the stellar object are examined,considering the Boulware-Deser 4D space-time as the external geometry.The physical properties of the stellar model are analyzed using parameters such as energy conditions,causality conditions,compactness,and redshift.
基金support from the Deanship of Scientific Research at King Khalid University,Saudi Arabia(RGP2/505/45)。
文摘Neodymium chromium oxide(NdCrO_(3))and NdCrO_(3)/graphene oxide(GO)nanocomposite were synthesized via sol-gel and co-precipitation techniques for being used in high-perfo rmance supercapacitors and for the possible application in ultraviolet(UV)materials.Herein the systematic synthesis approach was adopted,which enhances the optical and electrical properties of the grown wide band-gap composite nanomaterial.Structural characterization of the grown materials was attempted using X-ray diffraction(XRD)and scanning electron microscopy(SEM).Most importantly the electrochemical analysis of the grown samples was carried out by employing a glassy carbon electrode and 3 mol/L KOH electrolyte,which demonstrates significant improvements in a specific capacitance of approximately360 F/g,an energy density of approximately 18 Wh/kg,and a maximum power density of approximately 257 W/kg,respectively.Moreover,NdCrO_(3)/GO nanocomposite maintains a cyclic stability of 97.6%after4000 cycles.Photoluminescence(PL)spectroscopy confirms the wide bandgap nature of the NdCrO_(3)and NdCrO_(3)/GO nanocomposite,indicating its potential application in UVC devices.These findings emphasize the potential of the NdCrO_(3)/GO nanocomposite in advancing efficient energy storage solutions and the possibility of being used in UVC technology.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2504)。
文摘Accurate estimation of the Direction-of-Arrival(DoA)of incident plane waves is essential for modern wireless communication,radar,sonar,and localization systems.Precise DoA information enables adaptive beamforming,spatial filtering,and interference mitigation by steering antenna array beams toward desired sources while suppressing unwanted signals.Traditional one-dimensional Uniform Linear Arrays(ULAs)are limited to elevation angle estimation due to geometric constraints,typically within the range[0,π].To capture full spatial characteristics in environments with multipath and angular spread,joint estimation of both elevation and azimuth angles becomes necessary.However,existing 2D and 3D array geometries often entail increased hardware complexity and computational cost.This work proposes a novel and efficient framework for joint elevation and azimuth angle estimation using three spatially separated,parallel ULAs.The array configuration exploits spatial diversity and orthogonal projections to capture complete directional information with minimal structural overhead.A customized objective function based on the mean square error between measured and reconstructed array outputs is formulated to guide the estimation process.To solve the resulting non-convex optimization problem,three strategies are investigated:a global Genetic Algorithm(GA),a local Pattern Search(PS),and a hybrid GA-PS method that combines global exploration with local refinement.The proposed framework supports automatic pairing of elevation and azimuth angles,eliminating the need for manual post-processing.Extensive simulations validate the robustness,convergence,and accuracy of all three methods under varying signal-to-noise ratio conditions.Results confirm that the hybrid GA-PS approach achieves superior estimation performance and reduced computational complexity,making it well-suited for real-time and resource-constrained applications in next-generation sensing and communication systems.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2503)。
文摘The increasing demand for sustainable construction practices has led to growing interest in recycled aggregate concrete(RAC)as an eco-friendly alternative to conventional concrete.However,predicting its compressive strength remains a challenge due to the variability in recycled materials and mix design parameters.This study presents a robust machine learning framework for predicting the compressive strength of recycled aggregate concrete using feedforward neural networks(FFNN),Random Forest(RF),and XGBoost.A literature-derived dataset of 502 samples was enriched via interpolation-based data augmentation and modeled using five distinct optimization techniques within MATLAB’s Neural Net Fitting module:Bayesian Regularization,Levenberg-Marquardt,and three conjugate gradient variants—Powell/Beale Restarts,Fletcher-Powell,and Polak-Ribiere.Hyperparameter tuning,dropout regularization,and early stopping were employed to enhance generalization.Comparative analysis revealed that FFNN outperformed RF and XGBoost,achieving an R2 of 0.9669.To ensure interpretability,accumulated local effects(ALE)along with partial dependence plots(PDP)were utilized.This revealed trends consistent with the pre-existent domain knowledge.This allows estimation of strength from the properties of the mix without extensive lab testing,permitting designers to track the performance and sustainability trends in concrete mix designs while promoting responsible construction and demolition waste utilization.
基金supported and funded by theDeanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-RP23082).
文摘Fog computing is a key enabling technology of 6G systems as it provides quick and reliable computing,and data storage services which are required for several 6G applications.Artificial Intelligence(AI)algorithms will be an integral part of 6G systems and efficient task offloading techniques using fog computing will improve their performance and reliability.In this paper,the focus is on the scenario of Partial Offloading of a Task to Multiple Helpers(POMH)in which larger tasks are divided into smaller subtasks and processed in parallel,hence expediting task completion.However,using POMH presents challenges such as breaking tasks into subtasks and scaling these subtasks based on many interdependent factors to ensure that all subtasks of a task finish simultaneously,preventing resource wastage.Additionally,applying matching theory to POMH scenarios results in dynamic preference profiles of helping devices due to changing subtask sizes,resulting in a difficult-to-solve,externalities problem.This paper introduces a novel many-to-one matching-based algorithm,designed to address the externalities problem and optimize resource allocation within POMH scenarios.Additionally,we propose a new time-efficient preference profiling technique that further enhances time optimization in POMH scenarios.The performance of the proposed technique is thoroughly evaluated in comparison to alternate baseline schemes,revealing many advantages of the proposed approach.The simulation findings indisputably show that the proposed matching-based offloading technique outperforms existing methodologies in the literature,yielding a remarkable 52 reduction in task latency,particularly under high workloads.
基金supported by the National Natural Science Foundation of China(No.32271976,32371978)scientific and technological innovation funding of Fujian Agriculture and Forestry University(KFb22087,KFB23145).
文摘Due to the features and wide range of potential applications,cellulose ionogels are the subject of extensive research.Green celluloses have been employed as a three-dimensional skeleton network to restrict the ionic liquids(ILs)toward advanced ion-conductive ionogels.Diversiform cellulose ionogels with desirable perfor-mances,via physical/chemical reactions between cellulose and ILs,have been harvested,which have the po-tential to emerge as a bright star in the field of flexible electronics,such as sensors,electrolyte materials as power sources,and thermoelectric devices.Herein,a review regarding cellulose ionogels in terms of fundamental types of cellulose,formation strategies and mechanism,and principal properties is presented.Next,the diverse application prospects of cellulose ionogels in flexible electronics have been summarized.More importantly,the future challenges and advancing directions to be explored for cellulose ionogels are discussed.
基金supported by the Fundamental Research Funds for the Central Universities(Grant No.D5000230061)the Guangdong Basic and Applied Basic Research Foundation(Grant No.2025A1515011192).
文摘This study explores the magnetohydrodynamic(MHD)boundary layer flow of a water-based Boger nanofluid over a stretching sheet,with particular focus on the influences of nanoparticle diameter,nanolayer effects,and thermal radiation.The primary aim is to examine how variations in nanoparticle size and nanolayer thickness affect the hydrothermal behavior of the nanofluid.The model also incorporates the contributions of viscous dissipation and Joule heating within the heat transfer equation.The governing momentum and energy equations are converted into dimensionless partial differential equations(PDEs)using appropriate similarity variables and are numerically solved using the finite element method(FEM)implemented in MATLAB.Extensive validation of this method confirms its reliability and accuracy in numerical solutions.The findings reveal that increasing the diameter of copper nanoparticles significantly enhances the velocity profile,with a more pronounced effect observed at wider inter-particle spacings.A higher solvent volume fraction leads to decreased velocity and temperature distributions,while a greater relaxation time ratio improves velocity and temperature profiles due to the increased elastic response of the fluid.Moreover,enhancements in the magnetic parameter,thermal radiation,and Eckert number lead to an elevation in temperature profiles.Furthermore,higher nanolayer thickness reduces the temperature profile,whereas particle radius yields the opposite outcome.
文摘The Langrial iron ore deposits,located in the villages of Dubran and Darkot in Hazara,Pakistan,were evaluated using remote sensing,magnetic,and geochemical investigations.Data from ASTER,Landsat-8,and Sentinel-2 satellites were utilized and processed through techniques such as band ratio analysis,band compositing,and NDVI masking to reduce vegetation effects and to delineate various lithological units and mineralogical signatures within the study area.Magnetic anomalies revealed multiple levels of iron mineralization,with the hematite zone showing the most significant potential for high-grade iron ore.Geochemical analyses confirmed the presence of iron,along with minerals such as chromium,calcium,magnesium,and lead.In Dubran,mean iron concentrations are recorded at 370.94 mg/kg,whereas in Darkot,they reach up to 2052 mg/kg.The integration of remote sensing,magnetic,and geochemical data delineates key mineralized zones in the various parts of the study area.This research highlights the importance of combining geophysical and geochemical methodologies to refine mineral exploration efforts.The findings enhance our understanding of the Langrial iron ore deposits and highlight their economic potential for sustainable mining practices.This study will contribute to meeting the growing demand for iron ore resources and reducing Pakistan's reliance on imports,thereby promoting the sustainable development of local industries.
基金supported by the National Research Foundation(NRF),Republic of Korea,under project BK21 FOUR(4299990213939).
文摘Smart learning environments have been considered as vital sources and essential needs in modern digital education systems.With the rapid proliferation of smart and assistive technologies,smart learning processes have become quite convenient,comfortable,and financially affordable.This shift has led to the emergence of pervasive computing environments,where user’s intelligent behavior is supported by smart gadgets;however,it is becoming more challenging due to inconsistent behavior of Artificial intelligence(AI)assistive technologies in terms of networking issues,slow user responses to technologies and limited computational resources.This paper presents a context-aware predictive reasoning based formalism for smart learning environments that facilitates students in managing their academic as well as extra-curricular activities autonomously with limited human intervention.This system consists of a three-tier architecture including the acquisition of the contextualized information from the environment autonomously,modeling the system using Web Ontology Rule Language(OWL 2 RL)and Semantic Web Rule Language(SWRL),and perform reasoning to infer the desired goals whenever and wherever needed.For contextual reasoning,we develop a non-monotonic reasoning based formalism to reason with contextual information using rule-based reasoning.The focus is on distributed problem solving,where context-aware agents exchange information using rule-based reasoning and specify constraints to accomplish desired goals.To formally model-check and simulate the system behavior,we model the case study of a smart learning environment in the UPPAAL model checker and verify the desired properties in the model,such as safety,liveness and robust properties to reflect the overall correctness behavior of the system with achieving the minimum analysis time of 0.002 s and 34,712 KB memory utilization.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2504).
文摘Vehicular networks enable seamless connectivity for exchanging emergency and infotainment content.However,retrieving infotainment data from remote servers often introduces high delays,degrading the Quality of Service(QoS).To overcome this,caching frequently requested content at fog-enabled Road Side Units(RSUs)reduces communication latency.Yet,the limited caching capacity of RSUs makes it impractical to store all contents with varying sizes and popularity.This research proposes an efficient content caching algorithm that adapts to dynamic vehicular demands on highways to maximize request satisfaction.The scheme is evaluated against Intelligent Content Caching(ICC)and Random Caching(RC).The obtained results show that our proposed scheme entertains more contentrequesting vehicles as compared to ICC and RC,with 33%and 41%more downloaded data in 28%and 35%less amount of time from ICC and RC schemes,respectively.
文摘An expanding human population and technological progress demand clean and effective energy-storing systems.Within the realm of energy-storing devices,supercapacitors(SCs)have grabbed huge focus owing to their high-power density,unique cycling stability,and fast charging discharging capabilities.Electrode material has a prominent impact on the effectiveness of SCs.Several types of electrode materials have been used,encompassing varied metal oxides,activated carbon,conducting polymers,and MOFs.Metal organic frameworks(MOFs)are considered emerging electrode candidates,which could be ascribed to the tunable porosity,large surface areas,and designed morphology.This review shows a detailed analysis of various mono-,bi-,and tri-metallic MOFs along with derivatives in SC applications,their structural characteristics,and synthetic strategies.It also critically evaluates MOFs potential to boost the SC's energy density,power density,stability,and conductivity.Also,it underscores their significance in the establishment of future-oriented energy storage applications.
基金supported by Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia through the Researchers Supporting Project PNURSP2025R333.
文摘Diagnosing gastrointestinal tract diseases is a critical task requiring accurate and efficient methodologies.While deep learning models have significantly advanced medical image analysis,challenges such as imbalanced datasets and redundant features persist.This study proposes a novel framework that customizes two deep learning models,NasNetMobile and ResNet50,by incorporating bottleneck architectures,named as NasNeck and ResNeck,to enhance feature extraction.The feature vectors are fused into a combined vector,which is further optimized using an improved Whale Optimization Algorithm to minimize redundancy and improve discriminative power.The optimized feature vector is then classified using artificial neural network classifiers,effectively addressing the limitations of traditional methods.Data augmentation techniques are employed to tackle class imbalance,improving model learning and generalization.The proposed framework was evaluated on two publicly available datasets:Hyper-Kvasir and Kvasir v2.The Hyper-Kvasir dataset,comprising 23 gastrointestinal disease classes,yielded an impressive 96.0%accuracy.On the Kvasir v2 dataset,which contains 8 distinct classes,the framework achieved a remarkable 98.9%accuracy,further demonstrating its robustness and superior classification performance across different gastrointestinal datasets.The results demonstrate the effectiveness of customizing deep models with bottleneck architectures,feature fusion,and optimization techniques in enhancing classification accuracy while reducing computational complexity.
文摘The complex morphological,anatomical,physiological,and chemical mechanisms within the aging brain have been the hot topic of research for centuries.The aging process alters the brain structure that affects functions and cognitions,but the worsening of such processes contributes to the pathogenesis of neurodegenerative disorders,such as Alzheimer's disease.Beyond these observable,mild morphological shifts,significant functional modifications in neurotransmission and neuronal activity critically influence the aging brain.Understanding these changes is important for maintaining cognitive health,especially given the increasing prevalence of age-related conditions that affect cognition.This review aims to explore the age-induced changes in brain plasticity and molecular processes,differentiating normal aging from the pathogenesis of Alzheimer's disease,thereby providing insights into predicting the risk of dementia,particularly Alzheimer's disease.
文摘In recent years,there has been a concerted effort to improve anomaly detection tech-niques,particularly in the context of high-dimensional,distributed clinical data.Analysing patient data within clinical settings reveals a pronounced focus on refining diagnostic accuracy,personalising treatment plans,and optimising resource allocation to enhance clinical outcomes.Nonetheless,this domain faces unique challenges,such as irregular data collection,inconsistent data quality,and patient-specific structural variations.This paper proposed a novel hybrid approach that integrates heuristic and stochastic methods for anomaly detection in patient clinical data to address these challenges.The strategy combines HPO-based optimal Density-Based Spatial Clustering of Applications with Noise for clustering patient exercise data,facilitating efficient anomaly identification.Subsequently,a stochastic method based on the Interquartile Range filters unreliable data points,ensuring that medical tools and professionals receive only the most pertinent and accurate information.The primary objective of this study is to equip healthcare pro-fessionals and researchers with a robust tool for managing extensive,high-dimensional clinical datasets,enabling effective isolation and removal of aberrant data points.Furthermore,a sophisticated regression model has been developed using Automated Machine Learning(AutoML)to assess the impact of the ensemble abnormal pattern detection approach.Various statistical error estimation techniques validate the efficacy of the hybrid approach alongside AutoML.Experimental results show that implementing this innovative hybrid model on patient rehabilitation data leads to a notable enhance-ment in AutoML performance,with an average improvement of 0.041 in the R2 score,surpassing the effectiveness of traditional regression models.
基金supported by the MSIT(Ministry of Science and ICT),Republic of Korea,under the Convergence Security Core Talent Training Business Support Program(IITP-2025-RS-2023-00266605)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation).
文摘Surveillance systems can take various forms,but gait-based surveillance is emerging as a powerful approach due to its ability to identify individuals without requiring their cooperation.In the existing studies,several approaches have been suggested for gait recognition;nevertheless,the performance of existing systems is often degraded in real-world conditions due to covariate factors such as occlusions,clothing changes,walking speed,and varying camera viewpoints.Furthermore,most existing research focuses on single-person gait recognition;however,counting,tracking,detecting,and recognizing individuals in dual-subject settings with occlusions remains a challenging task.Therefore,this research proposed a variant of an automated gait model for occluded dual-subject walk scenarios.More precisely,in the proposed method,we have designed a deep learning(DL)-based dual-subject gait model(DSG)involving three modules.The first module handles silhouette segmentation,localization,and counting(SLC)using Mask-RCNN with MobileNetV2.The next stage uses a Convolutional block attention module(CBAM)-based Siamese network for frame-level tracking with a modified gallery setting.Following the last,gait recognition based on regionbased deep learning is proposed for dual-subject gait recognition.The proposed method,tested on Shri Mata Vaishno Devi University(SMVDU)-Multi-Gait and Single-Gait datasets,shows strong performance with 94.00%segmentation,58.36%tracking,and 63.04%gait recognition accuracy in dual-subject walk scenarios.
