检索结果-南通市图书馆

Deep Convolutional Neural Networks for Accurate Classification of Gastrointestinal Tract Syndromes

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第1期78卷 1207-1225页

作者： Zahid Farooq Khan muhammad Ramzan mudassar Raza muhammad Attique Khan Khalid Iqbal Taerang Kim Jae-Hyuk Cha Department of Computer Science COMSATS University IslamabadWah CampusWah47040Pakistan Department of Computer Science and Mathematics Lebanese American UniversityBeirut13-5053Lebanon Department of Computer Science HITEC UniversityTaxila47080Pakistan Department of Computer Science COMSATS University IslamabadAttock CampusAttock43600Pakistan Department of Computer Science Hanyang UniversitySeoul04763Korea

Accurate detection and classification of artifacts within the gastrointestinal(GI)tract frames remain a significant challenge in medical image *** science combined with artificial intelligence is advancing to automate the diagnosis and treatment of numerous *** to this is the development of robust algorithms for image classification and detection,crucial in designing sophisticated systems for diagnosis and *** study makes a small contribution to endoscopic image *** proposed approach involves multiple operations,including extracting deep features from endoscopy images using pre-trained neural networks such as Darknet-53 and ***,feature optimization utilizes the binary dragonfly algorithm(BDA),with the fusion of the obtained feature *** fused feature set is input into the ensemble subspace k nearest neighbors(ESKNN)*** Kvasir-V2 benchmark dataset,and the COMSATS University Islamabad(CUI)Wah private dataset,featuring three classes of endoscopic stomach images were *** assessments considered various feature selection techniques,including genetic algorithm(GA),particle swarm optimization(PSO),salp swarm algorithm(SSA),sine cosine algorithm(SCA),and grey wolf optimizer(GWO).The proposed model excels,achieving an overall classification accuracy of 98.25% on the Kvasir-V2 benchmark and 99.90% on the CUI Wah private *** approach holds promise for developing an automated computer-aided system for classifying GI tract syndromes through endoscopy images.

关键词： Feature fusion Darknet-53 Xception binary dragonfly algorithm ensemble

Bond-Slip Behavior of Steel Bar and Recycled Steel Fibre-Reinforced Concrete

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Renewable Materials 2024年第1期12卷 167-186页

作者： Ismail Shah Jing Li Nauman Khan Hamad R.Almujibah muhammad mudassar Rehman Ali Raza Yun Peng School of Architecture and Civil Engineering Yunnan Agricultural UniversityKunming650000China Institute of Mountain Hazards and Environment Chinese Academy of SciencesChengdu610041China Yunnan International Joint R&D Center of Smart Agriculture and Water Security Yunnan Agricultural UniversityKunming650201China School of Water Conservancy Yunnan Agricultural UniversityKunming650201China School of Civil Engineering Guizhou UniversityGuiyang550025China Department of Civil Engineering College of EngineeringTaif UniversityP.O.Box 11099Taif City21974Saudi Arabia School of Civil Engineering Zhengzhou UniversityZhengzhou450001China

Recycled steel fiber reinforced concrete is an innovative construction material that offers exceptional mechanical properties and *** is considered a sustainable material due to its low carbon footprint and environmental friendly *** study examines the key influencing factors that affect the behavior of this material,such as the steel fiber volume ratio,recycled aggregate replacement rate,concrete strength grade,anchorage length,and stirrup *** study investigates the bond failure morphology,bond-slip,and bond strength constitutive relationship of steel fiber recycled *** results show that the addition of steel fibers at 0.5%,1.0%,and 1.5%volume ratios can improve the ultimate bond strength of pull-out specimens by 9.05%,6.94%,and 5.52%,*** replacement rate of recycled aggregate has minimal effect on the typical bond strength of pull-out ***,the ultimate bond strengths of pull-out specimens with concrete strength grades C45 and C60 have improved compared to those with C30 *** specimens with longer anchorage lengths exhibit lower ultimate bond strength,with a reduction of 33.19%and 46.37%for anchorage lengths of 5D and 7D,respectively,compared to those without *** restraint of 1φ8 and 2φ8 improves the ultimate bond strength by 5.29%and 6.90%,*** fibers have a significant effect on the behavior of concrete after it cracks,especially during the stable expansion stage,crack instability expansion stage,and failure stage.

关键词： Recycled coarse aggregates(RCA) steel fiber bonding performance bond-slip environmental challenges

Recent Developments in Authentication Schemes Used in Machine-Type Communication Devices in Machine-to-Machine Communication: Issues and Challenges

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 93-115页

作者： Shafi Ullah Sibghat Ullah Bazai Mohammad Imran Qazi mudassar Ilyas Abid Mehmood muhammad Asim Saleem Muhmmad Aasim Rafique Arsalan Haider Ilyas Khan Sajid Iqbal Yonis Gulzar Kauser Hameed Department of Computer Engineering Balochistan University of Information TechnologyEngineering and Management SciencesQuetta87300Pakistan Department of Information Technology Balochistan University of Information TechnologyEngineeringand Management SciencesQuetta87300Pakistan Department of Information Systems College of Computer Sciences and Information TechnologyKing Faisal UniversityHofuf in Al-Ahsa31982Saudi Arabia Department of Management Information Systems College of Business AdministrationKing Faisal UniversityHofuf in Al-Ahsa31982Saudi Arabia Department of Software Engineering College of ComputingRiphah International UniversityFaisalabad44000Pakistan Department of Electrical Engineering Balochistan University of Information Technology Engineeringand Management SciencesQuetta87300Pakistan Department of Mathematics College of Science Al-ZulfiMajmaah UniversityAl-Majmaah11952Saudi Arabia

Machine-to-machine (M2M) communication plays a fundamental role in autonomous IoT (Internet of Things)-based infrastructure, a vital part of the fourth industrial revolution. Machine-type communication devices(MTCDs) regularly share extensive data without human intervention while making all types of decisions. Thesedecisions may involve controlling sensitive ventilation systems maintaining uniform temperature, live heartbeatmonitoring, and several different alert systems. Many of these devices simultaneously share data to form anautomated system. The data shared between machine-type communication devices (MTCDs) is prone to risk dueto limited computational power, internal memory, and energy capacity. Therefore, securing the data and devicesbecomes challenging due to factors such as dynamic operational environments, remoteness, harsh conditions,and areas where human physical access is difficult. One of the crucial parts of securing MTCDs and data isauthentication, where each devicemust be verified before data transmission. SeveralM2Mauthentication schemeshave been proposed in the literature, however, the literature lacks a comprehensive overview of current M2Mauthentication techniques and the challenges associated with them. To utilize a suitable authentication schemefor specific scenarios, it is important to understand the challenges associated with it. Therefore, this article fillsthis gap by reviewing the state-of-the-art research on authentication schemes in MTCDs specifically concerningapplication categories, security provisions, and performance efficiency.

关键词： Authentication cyber security internet of things machine-type communication devices machine-to-machine communication

Hydrological response under CMIP6 climate projection in Astore River Basin,Pakistan

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Mountain Science 2023年第8期20卷 2263-2281页

作者： Zeshan ALI mudassar IQBAL Ihsan Ullah KHAN muhammad Umer MASOOD muhammad UMER muhammad Usama Khan LODHI muhammad Atiq Ur Rehman TARIQ Centre of Excellence in Water Resources Engineering University of Engineering and TechnologyLahore 54890Pakistan International Water Management Institute Lahore 53700Pakistan Department of Civil Engineering University of TokyoTokyo 113-8656Japan College of Engineering IT&EnvironmentCharles Darwin UniversityDarwinNT 0810Australia

Climate change strongly influences the available water resources in a watershed due to direct linkage of atmospheric driving forces and changes in watershed hydrological *** how these climatic changes affect watershed hydrology is essential for human society and environmental *** Model Intercomparison Project phase 6(CMIP6)dataset of three GCM's(BCC-CSM2-MR,INM-CM5-0,and MPIESM1-2-HR)with resolution of 100 km has been analyzed to examine the projected changes in temperature and precipitation over the Astore catchment during *** correction method was used to reduce *** this study,statistical significance of trends was performed by using the Man-Kendall ***'s estimator determined the magnitude of the trend on both seasonal and annual scales at Rama Rattu and Astore ***-ESM1-2-HR showed better results with coefficient of determination(COD)ranging from 0.70-0.74 for precipitation and 0.90-0.92 for maximum and minimum temperature at Astore,Rama,and Rattu followed by INM-CM5-0 and *** of British Columbia Watershed model was used to attain the future hydrological series and to analyze the hydrological response of Astore River Basin to climate *** revealed that by the end of the 2070s,average annual precipitation is projected to increase up to 26.55%under the SSP1-2.6,6.91%under SSP2-4.5,and decrease up to 21.62%under the *** also showed considerable variability during summer and *** projected temperature showed an increasing trend that may cause melting of *** projected increase in temperature ranges from-0.66℃ to 0.50℃,0.9℃ to 1.5℃ and 1.18℃ to 2℃ under the scenarios of SSP1-2.6,SSP2-4.5 and SSP5-8.5,*** streamflows presented a slight increase by all *** streamflow was generated under SSP5-8.5 followed by SSP2-4.5 and *** snowmelt and groundwater contributions to streamflow have decreased whereas rainfa

关键词： GCMs UBCWM Astore River Climate Change Upper Indus Basin Bias Correction

Malaria Parasite Detection Using a Quantum-Convolutional Network

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第3期70卷 6023-6039页

作者： Javaria Amin muhammad Almas Anjum Abida Sharif mudassar Raza Seifedine Kadry Yunyoung Nam University of Wah Wah CanttPakistan National University of Technology(NUTECH) IslamabadPakistan COMSATS University Islamabad Vehari CampusVehariPakistan COMSATS University Islamabad Wah CampusPakistan Faculty of Applied Computing and Technology Noroff University CollegeKristiansandNorway Department of Computer Science and Engineering Soonchunhyang UniversityAsan31538Korea

Malaria is a severe illness triggered by parasites that spreads via mosquito *** underdeveloped nations,malaria is one of the top causes of mortality,and it is mainly diagnosed through ***-assisted malaria diagnosis is difficult owing to the fine-grained differences throughout the presentation of some uninfected and infected ***,in this study,we present a new idea based on the ensemble quantum-classical framework for malaria *** methods comprise three core steps:localization,segmentation,and *** the first core step,an improved FRCNN model is proposed for the localization of the infected malaria ***,the RGB localized images were converted into YCbCr channels to normalize the image intensity ***,the actual lesion region was segmented using a histogram-based color thresholding *** segmented images were employed for classification in two different *** the first method,a CNN model is developed by the selection of optimum layers after extensive experimentation,and the final computed feature vector is passed to the softmax layer for classification of the infection/non-infection of themicroscopicmalaria ***,a quantum-convolutionalmodel is employed for informative feature extraction from microscopicmalaria images,and the extracted feature vectors are supplied to the softmax layer for ***,classification results were analyzed from two different models and concluded that the quantum-convolutional model achieved maximum accuracy as compared to *** proposed models attain a precision rate greater than 90%,thereby proving that these models performed better than the existing models.

关键词： Quantum convolutional RGB YCbCr histogram Malaria

Improved Shark Smell Optimization Algorithm for Human Action Recognition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2023年第9期76卷 2667-2684页

作者： Inzamam Mashood Nasir mudassar Raza Jamal Hussain Shah muhammad Attique Khan Yun-Cheol Nam Yunyoung Nam Department of Computer Science COMSATS University IslamabadWah CampusWah Cantt47040Pakistan Department of Computer Science HITEC UniversityTaxilaPakistan Department of Architecture Joongbu UniversityGoyang10279South Korea Department of ICT Convergence Soonchunhyang UniversityAsan31538Korea

Human Action Recognition(HAR)in uncontrolled environments targets to recognition of different actions froma *** effective HAR model can be employed for an application like human-computer interaction,health care,person tracking,and video *** Learning(ML)approaches,specifically,Convolutional Neural Network(CNN)models had beenwidely used and achieved impressive results through feature *** accuracy and effectiveness of these models continue to be the biggest challenge in this *** this article,a novel feature optimization algorithm,called improved Shark Smell Optimization(iSSO)is proposed to reduce the redundancy of extracted *** proposed technique is inspired by the behavior ofwhite sharks,and howthey find the best prey in thewhole search *** proposed iSSOalgorithmdivides the FeatureVector(FV)into subparts,where a search is conducted to find optimal local features fromeach subpart of *** local optimal features are selected,a global search is conducted to further optimize these *** proposed iSSO algorithm is employed on nine(9)selected CNN *** CNN models are selected based on their top-1 and top-5 accuracy in ImageNet *** evaluate the model,two publicly available datasets UCF-Sports and Hollywood2 are selected.

关键词： Action recognition improved shark smell optimization convolutional neural networks machine learning

Liquidity connectedness in cryptocurrency market

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Financial Innovation 2022年第1期8卷 81-105页

作者： mudassar Hasan muhammad Abubakr Naeem muhammad Arif Syed Jawad Hussain Shahzad Xuan Vinh Vo Lahore Business School The University of LahoreLahorePakistan Smurfit Graduate School of Business University Col-lege DublinDublinIreland Institute of Business Research University of Economics Ho Chi Minh CityHo Chi Minh CityVietnam Department of Business Administration Shaheed Benazir Bhutto UniversityShaheed BenazirabadPakistan Montpellier Business School MontpellierFrance South Ural State University ChelyabinskRussian Federation Institute of Business Research and CFVG University of Economics Ho Chi Minh CityHo Chi Minh CityVietnam

We examine the dynamics of liquidity connectedness in the cryptocurrency *** use the connectedness models of Diebold and Yilmaz(Int J Forecast 28(1):57–66,2012)and Baruník and Křehlík(J Financ Econom 16(2):271–296,2018)on a sample of six major cryptocurrencies,namely,Bitcoin(BTC),Litecoin(LTC),Ethereum(ETH),Ripple(XRP),Monero(XMR),and *** static analysis reveals a moderate liquidity connectedness among our sample cryptocurrencies,whereas BTC and LTC play a significant role in connectedness magnitude.A distinct liquidity cluster is observed for BTC,LTC,and XRP,and ETH,XMR,and Dash also form another distinct liquidity *** frequency domain analysis reveals that liquidity connectedness is more pronounced in the short-run time horizon than the medium-and long-run time *** the short run,BTC,LTC,and XRP are the leading contributor to liquidity shocks,whereas,in the long run,ETH assumes this *** with the medium term,a tight liquidity clustering is found in the short and long *** time-varying analysis indicates that liquidity connectedness in the cryptocurrency market increases over time,pointing to the possible effect of rising demand and higher acceptability for this unique ***,more pronounced liquidity connectedness patterns are observed over the short and long run,reinforcing that liquidity connectedness in the cryptocurrency market is a phenomenon dependent on the time–frequency connectedness.

关键词： Liquidity Time-frequency connectedness Cryptocurrencies

Computational solution of an acid-mediated tumor-growth radial model under logistic growth regimes for normal and cancer cells

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

International Journal of Biomathematics 2023年第3期16卷 45-62页

作者： Farhan Khan mudassar Abbas Jorge E.Macias-Diaz muhammad Bilal Khan Safar M.Alghamdi Department of Mathematics and Computer Sciences University of Palermo 90133 PalermoItaly Department of Mathematics and Applications “R.Caccioppoli”University of Naples Federico iiVia Cintia 80126NaplesItaly Department of Mathematics School of Digital Technologies Tallinn University10120 TallinnEstonia Departamento de Matematicas y Fisica Universidad Autonoma de Aguascalientes AguascalientesAgs.20131Merico Department of Mathematics COMSATS University Islamabad Islambad44000 Pakistan Department of Mathematics and Statistics College of Science Taif University Taif 21944Saudi Arabia

Tumor invasion follows a complex mechanism which involves cell migration and *** study the processes in which primary and secondary metastases invade and damage the normal cells,mathematical models are often extremely *** this paper,we present a mathematical model of acid-mediated tumor growth consisting of radially symmetric reaction-diffusion *** assumption on the radial symmetry of the solutions is imposed here in view that tumors present spherical symmetry at the microscopic ***,we consider various empirical mechanisms which describe the propagation of tumors by considering cancer cells,normal cells,and the concentration of H+*** other assumptions,we suppose that these components follow logistictype growth ***,this is an important difference with respect to various other mathematical models for tumor growth available in the ***,we also add competition terms of normal and tumor cells *** carry out a balancing study of the equations of the model,and a numerical model is proposed to produce *** practical remarks derived from our assumptions are provided in the discussion of our simulations.

关键词： Cancer growth model radially symmetric system acid-mediated tumor growth computer simulations

Anomalous Situations Recognition in Surveillance Images Using Deep Learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2023年第7期76卷 1103-1125页

作者： Qurat-ul-Ain Arshad mudassar Raza Wazir Zada Khan Ayesha Siddiqa Abdul Muiz muhammad Attique Khan Usman Tariq Taerang Kim Jae-Hyuk Cha Department of Computer Science COMSATS University IslamabadWah Campus47040Pakistan Deptartment of Computer Science University of WahWah Cantt47040Pakistan Deptartment of Computer Science HITEC UniversityTaxila47080Pakistan Management Information System Department College of Business AdministrationPrince Sattam Bin Abdulaziz UniversityAl-Kharj 16278Saudi Arabia Department of Computer Science Hanyang UniversitySeoul04763Korea

Anomalous situations in surveillance videos or images that may result in security issues,such as disasters,accidents,crime,violence,or terrorism,can be identified through video anomaly ***,differentiat-ing anomalous situations from normal can be challenging due to variations in human activity in complex environments such as train stations,busy sporting fields,airports,shopping areas,military bases,care centers,*** learning models’learning capability is leveraged to identify abnormal situations with improved *** work proposes a deep learning architecture called Anomalous Situation Recognition Network(ASRNet)for deep feature extraction to improve the detection accuracy of various anomalous image *** proposed framework has five *** the first step,pretraining of the proposed architecture is performed on the CIFAR-100 *** the second step,the proposed pre-trained model and Inception V3 architecture are used for feature extraction by utilizing the suspicious activity recognition *** the third step,serial feature fusion is performed,and then the Dragonfly algorithm is utilized for feature optimization in the fourth ***,using optimized features,various Support Vector Machine(SVM)and K-Nearest Neighbor(KNN)based classification models are utilized to detect anomalous *** proposed framework is validated on the suspicious activity dataset by varying the number of optimized features from 100 to *** results show that the proposed method is effective in detecting anomalous situations and achieves the highest accuracy of 99.24%using cubic SVM.

关键词： Anomaly detection anomalous events anomalous behavior anomalous objects violence detection deep learning