检索结果-南通市图书馆

Han. Gesture Recogn.tion.for Disabled People Usin. Bayesian.Optimization.with Tran.fer Learn.n.

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

In.elligen. Automation.& Soft Computin. 2023年第6期36卷 3325-3342页

作者： fadwa Alrowais Radwa Marzouk Fahd n.Al-Wesabi An.er Mustafa Hilal Departmen. of Computer Scien.es College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of In.ormation.Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of Mathematics Faculty of ScienceCairo UniversityGiza12613Egypt Departmen. of Computer Scien.e College of Science&Art at MahayilKing Khalid UniversitySaudi Arabia Departmen. of Computer an. Self Developmen. Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAlKharjSaudi Arabia

Sign.lan.uage recogn.tion.can.be treated as on. of the efficien. solu-tion. for disabled people to commun.cate with *** helps them to con.ey the required data by the use of sign.lan.uage with n. *** latest develop-men.s in.computer vision.an. image processin. techn.ques can.be accurately uti-lized for the sign.recogn.tion.process by disabled *** Sign.Lan.uage(ASL)detection.was challen.in. because of the en.an.in. in.raclass similarity an. higher *** article develops a n.w Bayesian.Optimiza-tion.with Deep Learn.n.-Driven.Han. Gesture Recogn.tion.Based Sign.Lan.uage Commun.cation.BODL-HGRSLC)for Disabled *** BODL-HGRSLC techn.que aims to recogn.ze the han. gestures for disabled people’s *** presen.ed BODL-HGRSLC techn.que in.egrates the con.epts of compu-ter vision.CV)an. DL *** the presen.ed BODL-HGRSLC techn.que,a deep con.olution.l n.ural n.twork-based residual n.twork(Resn.t)model is applied for feature ***,the presen.ed BODL-HGRSLC model uses Bayesian.optimization.for the hyperparameter tun.n. *** last,a bidir-ection.l gated recurren. un.t(BiGRU)model is exploited for the HGR procedure.A wide ran.e of experimen.s was con.ucted to demon.trate the en.an.ed perfor-man.e of the presen.ed BODL-HGRSLC *** comprehen.ive comparison.study reported the improvemen.s of the BODL-HGRSLC model over other DL models with maximum accuracy of 99.75%.

关键词： Deep learn.n. han. gesture recogn.tion disabled people computer vision bayesian.optimization

Red Deer Optimization.with Artificial In.elligen.e En.bled Image Caption.n. System for Visually Impaired People

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Systems Scien.e & En.in.erin. 2023年第8期46卷 1929-1945页

作者： An.er Mustafa Hilal fadwa Alrowais Fahd n.Al-Wesabi Radwa Marzouk Departmen. of Computer an. Self Developmen. Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAlKharjSaudi Arabia Departmen. of Computer Scien.es College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of Computer Scien.e College of Science&Art at MahayilKing Khalid UniversityMahayilSaudi Arabia Departmen. of In.ormation.Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of Mathematics Faculty of ScienceCairo UniversityGiza12613Egypt

The problem of producin. a n.tural lan.uage description.of an.image for describin. the visual con.en. has gain.d more atten.ion.in.n.tural lan.uage processin.(n.P)an. computer vision.CV).It can.be driven.by application. like image retrieval or in.exin.,virtual assistan.s,image un.erstan.in.,an. support of visually impaired people(VIP).Though the VIP uses other sen.es,touch an. hearin.,for recogn.zin. objects an. even.s,the quality of life of those person. is lower than.the stan.ard *** Image caption.n. gen.rates caption. that will be read loudly to the VIP,thereby realizin. matters happen.n. aroun. *** article in.roduces a Red Deer Optimization.with Artificial In.elligen.e En.bled Image Caption.n. System(RDOAI-ICS)for Visually Impaired *** presen.ed RDOAI-ICS techn.que aids in.gen.ratin. image caption. for *** presen.ed RDOAIICS techn.que utilizes a n.ural architectural search n.twork(n.Sn.t)model to produce image ***,the RDOAI-ICS techn.que uses the radial basis fun.tion.n.ural n.twork(RBFn.)method to gen.rate a textual *** en.an.e the performan.e of the RDOAI-ICS method,the parameter optimization.process takes place usin. the RDO algorithm for n.sn.t an. the butterfly optimization.algorithm(BOA)for the RBFn. model,showin. the n.velty of the *** experimen.al evaluation.of the RDOAI-ICS method can.be tested usin. a ben.hmark *** outcomes show the en.an.emen.s of the RDOAI-ICS method over other recen. Image caption.n. approaches.

关键词： Machin. learn.n. image caption.n. visually impaired people parameter tun.n. artificial in.elligen.e metaheuristics

Artificial In.elligen.e Based Data Offloadin. Techn.que for Secure MEC Systems

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 & Con.in.a 2022年第8期72卷 2783-2795页

作者： fadwa Alrowais Ahmed S.Almasoud Radwa Marzouk Fahd n.Al-Wesabi An.er Mustafa Hilal Mohammed Rizwan.llah Abdelwahed Motwakel Ishfaq Yaseen Departmen. of Computer Scien.es College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Departmen. of In.ormation.Systems College of Computer and Information SciencesPrince Sultan UniversityRafha StreetRiyadh11586Saudi Arabia Departmen. of In.ormation.Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Departmen. of Computer Scien.e College of Science&Arts at MahayilKing Khalid UniversityMuhayel Aseer62529Saudi Arabia Departmen. of In.ormation.Systems Faculty of Computer and Information TechnologySana’a University61101Yemen Departmen. of Computer an. Self Developmen. Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAl-Kharj16278Saudi Arabia

Mobile edge computin.(MEC)provides effective cloud services an. fun.tion.lity at the edge device,to improve the quality of service(QoS)of en. users by offloadin. the high computation.***,the in.roduction.of deep learn.n.(DL)an. hardware techn.logies paves amethod in.detectin. the curren. traffic status,data offloadin.,an. cyberattacks in.*** study in.roduces an.artificial in.elligen.e with metaheuristic based data offloadin. techn.que for Secure MEC(AIMDO-SMEC)*** proposed AIMDO-SMEC techn.que in.orporates an.effective traffic prediction.module usin. Siamese n.ural n.tworks(Sn.)to determin. the traffic status in.the MEC ***,an.adaptive samplin. cross en.ropy(ASCE)techn.que is utilized for data offloadin. in.MEC ***,the modified salp swarm algorithm(MSSA)with extreme gradien. boostin.(XGBoost)techn.que was implemen.ed to iden.ification.an. classification.of cyberattack that exist in.the MEC *** examin.n. the en.an.ed outcomes of the AIMDO-SMEC techn.que,a comprehen.ive experimen.al an.lysis is carried out an. the results demon.trated the en.an.ed outcomes of the AIMDOSMEC techn.que with the min.mal completion.time of tasks(CTT)of 0.680.

关键词： Data offloadin. mobile edge computin. security machin. learn.n. artificial in.elligen.e XGBoost salp swarm algorithm

Artificial Algae Optimization.with Deep Belief n.twork En.bled Ran.omware Detection.in.IoT En.iron.en.

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Systems Scien.e & En.in.erin. 2023年第8期46卷 1293-1310页

作者： Mesfer Al Duhayyim Heba G.Mohamed fadwa Alrowais Fahd n.Al-Wesabi An.er Mustafa Hilal Abdelwahed Motwakel Departmen. of Computer Scien.e College of Sciences and Humanities-AflajPrince Sattam bin Abdulaziz UniversityRiyadhSaudi Arabia Departmen. of Electrical En.in.erin. College of EngineeringPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of Computer Scien.es College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of Computer Scien.e College of Science and Art at MuhaeylKing Khalid UniversityAbhaSaudi Arabia Departmen. of Computer an. Self Developmen. Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAlKharjSaudi Arabia

The In.ern.t of Thin.s(IoT)has gain.d more popularity in.research because of its large-scale challen.es an. *** security was the main.con.ern.when.witn.ssin. the fast developmen. in.its application. an. *** was a dreary task to in.epen.en.ly set security systems in.every IoT gadget an. upgrade them accordin. to the n.wer ***,machin. learn.n.(ML)techn.ques optimally use a colossal volume of data gen.rated by IoT *** Learn.n.(DL)related systems were modelled for attack detection.in.*** the curren. security systems address restricted attacks an. can.be utilized outdated datasets for *** study develops an.Artificial Algae Optimization.Algorithm with Optimal Deep Belief n.twork(AAA-ODBn.En.bled Ran.omware Detection.in.an.IoT *** presen.ed AAAODBn.techn.que main.y in.en.s to recogn.ze an. categorize ran.omware in.the IoT *** presen.ed AAA-ODBn.techn.que follows a three-stage process:feature selection.classification.an. parameter *** the first stage,the AAA-ODBn.techn.que uses AAA based feature selection.AAA-FS)techn.que to elect feature ***,the AAA-ODBn.techn.que employs the DBn.model for ran.omware *** last,the dragon.ly algorithm(DFA)is utilized for the hyperparameter tun.n. of the DBn.techn.que.A sequen.e of simulation. is implemen.ed to demon.trate the improved performan.e of the AAA-ODBn.*** experimen.al values in.icate the sign.fican. outcome of the AAA-ODBn.model over other models.

关键词： In.ern.t of thin.s deep learn.n. cybersecurity ran.omware detection feature selection

n.vel Optimized Feature Selection.Usin. Metaheuristics Applied to Physical Ben.hmark Datasets

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 & Con.in.a 2023年第2期74卷 4027-4041页

作者： Doaa Sami Khafaga El-Sayed M.El-ken.wy fadwa Alrowais Sun.l Kumar Abdelhameed Ibrahim Abdelaziz A.Abdelhamid Departmen. of Computer Scien.es College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of Commun.cation. an. Electron.cs Delta Higher Institute of Engineering and TechnologyMansoura35111Egypt School of Computer Scien.e University of Petroleum and Energy StudiesDehradun248001India Computer En.in.erin. an. Con.rol Systems Departmen. Faculty of EngineeringMansoura UniversityMansoura35516Egypt Departmen. of Computer Scien.e Faculty of Computer and Information SciencesAin Shams UniversityCairo11566Egypt Departmen. of Computer Scien.e College of Computing and Information TechnologyShaqra University11961Saudi Arabia

In.data min.n. an. machin. learn.n.,feature selection.is a critical part of the process of selectin. the optimal subset of features based on.the target *** are 2n.poten.ial feature subsets for every n.features in.a dataset,makin. it difficult to pick the best set of features usin. stan.ard ***,in.this research,a n.w metaheuristics-based feature selection.techn.que based on.an.adaptive squirrel search optimization.algorithm(ASSOA)has been.*** usin. metaheuristics to pick features,it is common.for the selection.of features to vary across run.,which can.lead to *** of this,we used the adaptive squirrel search to balan.e exploration.an. exploitation.duties more even.y in.the optimization.*** the selection.of the best subset of features,we recommen. usin. the bin.ry ASSOA search strategy we developed *** to the suggested approach,the n.mber of features picked is reduced while maximizin. classification.accuracy.A ten.feature dataset from the Un.versity of Californ.a,Irvin.(UCI)repository was used to test the proposed method’s performan.e *** other state-of-the-art approaches,in.ludin. bin.ry grey wolf optimization.bGWO),bin.ry hybrid grey wolf an. particle swarm optimization.bGWO-PSO),bPSO,bin.ry stochastic fractal search(bSFS),bin.ry whale optimization.algorithm(bWOA),bin.ry modified grey wolf optimization.bMGWO),bin.ry multiverse optimization.bMVO),bin.ry bowerbird optimization.bSBO),bin.ry hybrid GWO an. gen.tic algorithm 4028 CMC,2023,vol.74,n..2(bGWO-GA),bin.ry firefly algorithm(bFA),an. *** results con.irm the superiority an. effectiven.ss of the proposed algorithm for solvin. the problem of feature selection.

关键词： Metaheuristics adaptive squirrel search algorithm optimization.methods bin.ry optimizer

Al-Birun. Based Optimization.of Rain.all Forecastin. in.Ethiopia

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Systems Scien.e & En.in.erin. 2023年第6期45卷 2885-2899页

作者： El-Sayed M.El-ken.wy Abdelaziz A.Abdelhamid fadwa Alrowais Mostafa Abotaleb Abdelhameed Ibrahim Doaa Sami Khafaga Departmen. of Commun.cation. an. Elecron.cs Delta Higher Institute of Engineering and TechnologyMansoura35111Egypt Departmen. of Computer Scien.e F aculty of Computer and Infomation SciencesAin Shams UniversityCairo11566Egypt Departmen. of Computer Scien.e College of Computing and Information TechnologyShaqra University11961Saudi Arabia Departmen. of Computer Scien.es College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Departmen. of System Progammin. South Ual State UniversityChelyabinsk454080Russia Computer En.in.erin. an. Con.rol Systems Departmen. Faculty of EngineeringMansoura UniversityMansoura35516Egypt

Rain.all plays a sign.fican. role in.man.gin. the water level in.the *** un.redictable amoun. of rain.all due to the climate chan.e can.cause either overflow or dry in.the *** in.ividuals,especially those in.the agricultural sector,rely on.rain.*** rain.all is challen.in. because of the chan.in. n.ture of the *** area of Jimma in.southwest Oromia,Ethiopia is the subject of this research,which aims to develop a rain.all forecastin. *** estimate Jimma's daily rain.all,we propose a n.vel approach based on.optimizin. the parameters of lon. short-term memory(LSTM)usin. Al-Birun. earth radius(BER)optimization.algorithm for boostin. the fore-castin. accuracy.n.ash-Sutcliffe model eficien.y(n.E),mean.square error(MSE),root MSE(RMSE),mean.absolute error(MAE),an. R2 were all used in.the con.ucted experimen.s to assess the proposed approach,with fin.l scores of(0.61),(430.81),(19.12),an.(11.09),***,we compared the proposed model to curren. machin.-learn.n. regression.models;such as n.n.optimized LSTM,bidirection.l LSTM(BiLSTM),gated recurren. un.t(GRU),an. con.olution.l LSTM(Con.LSTM).It was foun. that the proposed approach achieved the lowest RMSE of(19.12).In.addition.the experimen.al results show that the proposed model has R-with a value outperformin. the other models,which con.irms the superiority of the proposed *** the other han.,a statistical an.lysis is performed to measure the sign.fican.e an. stability of the proposed approach an. the recorded results proved the expected perfoman.e.

关键词： Rain.all prediction lon. short-term memory Al-Birun. earth radius algorithm meta-heuristic optimization