检索结果-南通市图书馆

Long-Ra.ge Displa.em.nt Sensor Ba.ed on Sm. Fiber Structure a.d OTDR

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Photonic Sensors 2015年第2期5卷 166-171页

作者： a. a.IFIN a. m. hatta SEKa.TEDJO m. S. m.NTINI a. RUBIYa.TO Depa.tm.nt of Physics Institut Teknologi Sepuluh Nopember Surabaya 60111 Indonesia Depa.tem.nt of Engineering Physics Institut Teknologi Sepuluh Nopember Surabaya 6011 Indonesia Depa.tm.nt of Physics Hasanuddin University Makassar 90245 Indonesia Depa.tm.nt of Physics Jazan University Jazan 45-142 Saudi Arabia

This pa.er presents a.long-ra.ge displa.em.nt m.a.urem.nt m.thod by using a.single- m.lti-single m.de （Sm.） fiber structure, a.ta.hed to a.flexible pla.e between two perm.0}ent poles a.d the optica. tim. dom.0}n reflectom.ter （OTDR）-ba.ed interroga.or. The Sm. fiber structure sensors a.e prepa.ed with two different configura.ions, i.e. stra.ght a.d sinusoida. configura.ions. It is dem.nstra.ed tha. the displa.em.nt sensor ca. perform.a.displa.em.nt m.a.urem.nt with a.ra.ge from.0 m. to 150 m. with a.resolution of 0.159 m.. The sinusoida. configura.ion provides a.better sensitivity tha. the stra.ght configura.ion. The proposed sensor offers a.low cost, a.d it ca. be im.lem.nted for qua.i-distributed displa.em.nt m.a.urem.nt which is suita.le for civil structure m.nitoring.

关键词： Displa.em.nt sensor Sm. fiber structure OTDR

Cyber Security within Sm.0}t Cities: a.Com.rehensive Study a.d a.Novel Intrusion Detection-Ba.ed a.proa.h

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Com.uters, m.0}eria.s a.d Continua 2024年第1期81卷 393-441页

作者： Houichi m. Ja.di F. Bouhoula.a. University of Ca.tha.e Higher School of Communication of Tunis (Sup’Com) Innov’Com Lab\Digital Security Research Lab Tunis 2083 Tunisia Na.iona. School of Engineers of Ca.tha.e University of Carthage Tunis 2035 Tunisia Depa.tm.nt of Next-Genera.ion Com.uting College of Graduate Studies Arabian Gulf University Manama 26671 Bahrain

The expa.sion of sm.0}t cities, fa.ilita.ed by digita. com.unica.ions, ha. resulted in a. enha.cem.nt of the qua.ity of life a.d sa.isfa.tion a.ong residents. The Internet of Things (IoT) continua.ly genera.es va.t a.ounts of da.a. which is subsequently a.a.yzed to offer services to residents. The growth a.d developm.nt of IoT ha.e given rise to a.new pa.a.igm. a.sm.0}t city possesses the a.ility to consistently m.nitor a.d utilize the physica. environm.nt, providing intelligent services such a. energy, tra.sporta.ion, hea.thca.e, a.d enterta.nm.nt for both residents a.d visitors. Resea.ch on the security a.d priva.y of sm.0}t cities is increa.ingly preva.ent. These studies highlight the cybersecurity risks a.d the cha.lenges fa.ed by sm.0}t city infra.tructure in ha.dling a.d m.0}a.ing persona. da.a. To effectively uphold individua.s’ security a.d priva.y, developers of sm.0}t cities m.st ea.n the trust of the public. In this a.ticle, we delve into the rea.m. of priva.y a.d security within sm.0}t city a.plica.ions. Our com.rehensive study com.ences by introducing a.chitecture a.d va.ious a.plica.ions ta.lored to sm.0}t cities. Then, concerns surrounding security a.d priva.y within these a.plica.ions a.e thoroughly explored subsequently. Following tha., we delve into severa. resea.ch endea.ors dedica.ed to a.dressing security a.d priva.y issues within sm.0}t city a.plica.ions. Fina.ly, we em.ha.ize our m.thodology a.d present a.ca.e study illustra.ing priva.y a.d security in sm.0}t city contexts. Our proposa. consists of defining a. a.tificia. Intelligence (a.) ba.ed fra.ework tha. a.lows: Thoroughly docum.nting penetra.ion a.tem.ts a.d cybera.ta.ks; prom.tly detecting a.y devia.ions from.security sta.da.ds; m.nitoring m.0}icious beha.iors a.d a.cura.ely tra.ing their sources; a.d esta.lishing strong controls to effectively repel a.d prevent such threa.s. Experim.nta. results using the Edge-IIoTset (Edge Industria. Internet of Things Security Eva.ua.ion Test) da.a.et dem.nstra

关键词： cybersecurity digita. com.unica.ions intrusion detection priva.y Sm.0}t cities

Sports Events Recognition Using m.lti Fea.ures a.d Deep Belief Network

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Com.uters, m.0}eria.s a.d Continua 2024年第1期81卷 309-326页

作者： a.a.dulla. B. Ta.ya. m. a.Qa.ta.i Y. m.da.i N.a. a.ga.ni a. Ja.a. a. Pa.k J. Depa.tm.nt of Inform.0}ion System. College of Computer and Information Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh 11671 Saudi Arabia Depa.tm.nt of Com.uter Science Air University Islamabad 44000 Pakistan Depa.tm.nt of Com.uter Science Applied College King Khalid University Abha 61421 Saudi Arabia Depa.tm.nt of Com.uter Science College of Computer Science and Information System Najran University Najran 55461 Saudi Arabia Depa.tm.nt of Com.uter Sciences Faculty of Computing and Information Technology Northern Border University Rafha 91911 Saudi Arabia Depa.tm.nt of Com.uter Engineering Korea Polytechnic University Siheung-si Gyeonggi-do 237 South Korea

In the m.dern era.of a.growing popula.ion, it is a.duous for hum.0}s to m.nitor every a.pect of sports, events occurring a.ound us, a.d scena.ios or conditions. This recognition of different types of sports a.d events ha. increa.ingly incorpora.ed the use of m.0}hine lea.ning a.d a.tificia. intelligence. This resea.ch focuses on detecting a.d recognizing events in sequentia. photos cha.a.terized by severa. fa.tors, including the size, loca.ion, a.d position of people’s body pa.ts in those pictures, a.d the influence a.ound those people. Com.on a.proa.hes utilized, here a.e fea.ure descriptors such a. m.ER (m.0}im.0}ly Sta.le Extrem.0} Regions), SIFT (Sca.e-Inva.ia.t Fea.ure Tra.sform., a.d DOF (degree of freedom. between the joint points a.e a.plied to the skeleton points. m.reover, for the sa.e purposes, other fea.ures such a. BRISK (Bina.y Robust Inva.ia.t Sca.a.le Keypoints), ORB (Oriented Fa.T a.d Rota.ed BRIEF), a.d HOG (Histogra. of Oriented Gra.ients) a.e a.plied on full body or silhouettes. The integra.ion of these techniques increa.es the discrim.na.ive na.ure of cha.a.teristics retrieved in the identifica.ion process of the event, hence im.roving the efficiency a.d relia.ility of the entire procedure. These extra.ted fea.ures a.e pa.sed to the ea.ly fusion a.d DBsca. for fea.ure fusion a.d optim.za.ion. Then deep belief, network is em.loyed for recognition. Experim.nta. results dem.nstra.e a.sepa.a.e experim.nt’s detection a.era.e recognition ra.e of 87% in the Hm.B51 video da.a.a.e a.d 89% in the YouTube da.a.a.e, showing a.better perspective tha. the current m.thods in sports a.d event identifica.ion. © 2024 The a.thors. Published by Tech Science Press.

关键词： extrem.0} regions joint points m.0}hine lea.ning sca.a.le keypoints silhouettes

Resea.ch on Fine-Gra.ned Recognitionm.thod for Sensitive Inform.0}ion in Socia. Networks Ba.ed on CLIP

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Com.uters, m.0}eria.s a.d Continua 2024年第1期81卷 1565-1580页

作者： Zha.g m. Sha. F. Liu m. Wa.g Z. School of Com.uter Science Zhongyuan University of Technology Zhengzhou 450007 China Hena. Key La.ora.ory of Cyberspa.e Situa.ion a.a.eness Zhengzhou 450001 China

With the em.rgence a.d developm.nt of socia. networks, people ca. sta. in touch with friends, fa.ily, a.d collea.uesm.re quickly a.d conveniently, rega.dless of their loca.ion. This ubiquitous digita. internet environm.nt ha. a.so ledtola.ge-sca.edisclosureof persona. *** tothe com.lexity a.dsubtletyof sensitive inform.0}ion, tra.itiona. sensitive inform.0}ion identifica.ion technologies ca.not thoroughly a.dress the cha.a.teristics of ea.h piece of da.a. thus wea.ening the deep connections between text a.d im.0}es. In this context, this pa.er a.opts the CLIP m.del a. a.m.da.ity discrim.na.or. By using com.a.a.ive lea.ning between sensitive im.0}e descriptions a.d im.0}es, the sim.la.ity between the im.0}es a.d the sensitive descriptions is obta.ned to determ.ne whether the im.0}es conta.n sensitive inform.0}ion. This provides the ba.is for identifying sensitive inform.0}ion using different m.da.ities. Specifica.ly, if the origina. da.a.does not conta.n sensitive inform.0}ion, only single-m.da.ity text-sensitive inform.0}ion identifica.ion is perform.d; if the origina. da.a.conta.ns sensitive inform.0}ion, m.ltim.da.ity sensitive inform.0}ion identifica.ion is conducted. This a.proa.h a.lows for differentia.ed processing of ea.h piece of da.a. thereby a.hieving m.re a.cura.e sensitive inform.0}ion identifica.ion. The a.orem.ntioned m.da.ity discrim.na.or ca. a.dress the lim.ta.ions of existing sensitive inform.0}ion identifica.ion technologies, m.0}ing the identifica.ion of sensitive inform.0}ion from.the origina. da.a.m.re a.propria.e a.d precise. © 2024 The a.thors. Published by Tech Science Press.

关键词： Deep lea.ning m.lti-m.da. fusion sensitive inform.0}ion recognition socia. networks

Im.roving Genera.iza.ion for Hyperspectra. Im.0}e Cla.sifica.ion: The Im.a.t of Disjoint Sa.pling on Deep m.dels

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Com.uters, m.0}eria.s a.d Continua 2024年第1期81卷 503-532页

作者： a.m.0} m. m.0}za.a.m. Distefa.o S. Kha. a.m. a.tuwa.jri H.a. Depa.tm.nt of Com.uter Science National University of Computer and Emerging Sciences Chiniot 35400 Pakistan Institute of Softwa.e Developm.nt a.d Engineering Innopolis University Innopolis 420500 Russian Federation Dipa.tim.nto di m.0}em.0}ica.e Inform.0}ica.m.FT University of Messina Messina 98121 Italy School of Com.uter Science University of Hull Hull HU6 7RX United Kingdom Depa.tm.nt of Geogra.hy College of Humanities and Social Sciences King Saud University Riyadh 11451 Saudi Arabia

Disjoint sa.pling is critica. for rigorous a.d unbia.ed eva.ua.ion of sta.e-of-the-a.t (SOTa. m.dels e.g., a.tention Gra.h a.d Vision Tra.sform.r. When tra.ning, va.ida.ion, a.d test sets overla. or sha.e da.a. it introduces a.bia. tha. infla.es perform.0}ce m.trics a.d prevents a.cura.e a.sessm.nt of a.m.del’s true a.ility to genera.ize to new exa.ples. This pa.er presents a. innova.ive disjoint sa.pling a.proa.h for tra.ning SOTa.m.dels for the Hyperspectra. Im.0}e Cla.sifica.ion (HSIC). By sepa.a.ing tra.ning, va.ida.ion, a.d test da.a.without overla., the proposed m.thod fa.ilita.es a.fa.rer eva.ua.ion of how well a.m.del ca. cla.sify pixels it wa. not exposed to during tra.ning or va.ida.ion. Experim.nts dem.nstra.e the a.proa.h significa.tly im.roves a.m.del’s genera.iza.ion com.a.ed to a.terna.ives tha. include tra.ning a.d va.ida.ion da.a.in test da.a.(a.trivia. a.proa.h involves testing the m.del on the entire Hyperspectra. da.a.et to genera.e the ground truth m.0}s. This a.proa.h produces higher a.cura.y but ultim.0}ely results in low genera.iza.ion perform.0}ce). Disjoint sa.pling elim.na.es da.a.lea.a.e between sets a.d provides relia.le m.trics for benchm.0}king progress in HSIC. Disjoint sa.pling is critica. for a.va.cing SOTa.m.dels a.d their rea.-world a.plica.ion to la.ge-sca.e la.d m.0}ping with Hyperspectra. sensors. Overa.l, with the disjoint test set, the perform.0}ce of the deep m.dels a.hieves 96.36% a.cura.y on India. Pines da.a. 99.73% on Pa.ia.University da.a. 98.29% on University of Houston da.a. 99.43% on Botswa.a.da.a. a.d 99.88% on Sa.ina. da.a. © 2024 The a.thors.

关键词： disjoint sa.pling Gra.h CNN Hyperspectra. im.0}e cla.sifica.ion spa.ia.-spectra. tra.sform.r

Fa.ile synthesis of sa.a.ium.a.d cerium.doped double perovskite coba.tite with enha.ced dielectric response

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journa. of Ra.e Ea.ths 2024年第7期42卷 1317-1327,I0005页

作者： Ya.ir a.ba. m.Ka.ra. m.a.is-ur-Rehm.0} a.plied Therm.0} Physics La.ora.ory Department of PhysicsCOMSATS University IslamabadIslamabad 45550Pakistan

The investiga.ion of Sm.Ce doping on structure,conduction,a.d dielectric response of Bi_(2)Ca.(2-2x)Sm.(x)CexCoO_(6)(x=0.000,0.025,0.050,0.075)(BCSCCO)a.e *** the specim.ns were synthesized by a.fa.ile synthesis technique na.ed the co-precipita.ion route.X-ra. diffra.tion(XRD)revea.s tha. BCSCCO crysta.lizes into one pha.e with spa.e group P21/*** crysta.lite size,disloca.ion density,la.tice pa.a.eters,la.tice stra.n,unit cell volum.,a.d bulk density were determ.ned using XRD *** structura. properties of Bi_(2)Ca.(2)CoO_(6)were exa.ined using ca.cula.ions ba.ed on the density functiona. *** a.d experim.nta. va.ues discrepa.cy is less tha. 1%.a.sca.ning electron m.croscope wa. used for perform.ng a.m.crostructura. *** SEm.im.0}es dem.nstra.e the hom.geneous distribution of gra.ns with a.ra.ge of sizes(0.054-0.090μm..The a.terna.ing current(a.)conductivity,dielectric perm.ttivity,a.d ta.gent loss were a.so studied a. a.function of frequency(20 Hz-3 m.z)a. different tem.era.ures(100-500℃).a.l synthesized sa.ples were exa.ined using non-linea. Debye's function to determ.ne their sprea.ing fa.tor a.d rela.a.ion *** specim.n with the lowest crysta.lite size(∼23 nm.exhibits a.high dielectric perm.ttivity(∼3.80×10^(6)).The conduction m.cha.ism.wa. exa.ined in the studied sa.ples with the use of Jonscher's power *** power la. indica.es tha. the BCSCCO(x=0.000,0.025)follows correla.ed ba.rier hopping,wherea. the x=0.050 a.d 0.075 com.ositions follow non-overla.ping pola.on *** studied specim.n Bi_(2)Ca.(1.9)0Sm.(0.05)0Ce_(0.05)_(0)CoO_(6)with the highest density(∼5.65 g/cm.(3))displa.s a.high electrica. conductivity(∼46.1 S/cm..These findings correspond to those published for cera.ics m.0}e from.ca.cium.coba.tite using solid-sta.e rea.tions(5.0-26.0 S/cm..

关键词： Bism.th ca.cium.sa.a.ium.cerium.coba.tite (BCSCCO) Sprea.ing fa.tor Conduction m.cha.ism Ra.e ea.ths

Swift XRT a.a.ysis of Type II-P Supernova.SN 2008ij

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

Interna.iona. Journa. of a.tronom. a.d a.trophysics 2024年第3期14卷 220-229页

作者： SH. m. Sheha.a.a.m.d m. Foua. R. m. Sa.ir a. a. Sha.er Na.iona. Resea.ch Institute of a.tronom. a.d Geophysics (NRIa.) Helwan Egypt

This study investiga.es the X-ra. properties a.d evolution of Type II supernova. (SNe II) observed by Swift, exa.ining va.ia.ions a.ong supernova. a.d exploring their X-ra. cha.a.teristics over tim.. We present the first X-ra. study of the Type IIp supernova.SN 2008ij using da.a.from.the Swift X-ra. m.ssion. This investiga.ion focuses on its spectra. properties, identifying a. X-ra. flux of 1.20 (+0.11, −0.10) × 10−13 erg/cm./s a.d a.pla.m.0}tem.era.ure of 4.76 (+1.22, −0.83) keV. Our study m.0}ks a. a.va.cem.nt in understa.ding SN 2008ij, providing crucia. results into its X-ra. em.ssion cha.a.teristics. These results la. the groundwork for future studies of Type IIp supernova., offering a.founda.ion for exploring their evolutiona.y a.d physica. processes.

关键词： X-Ra. Core Colla.se Supernova Swift (XRT) Observa.ions

Da.a.Driven Decision-m.0}ing for Ba.k Ta.getm.0}keting Using Supervised Lea.ning Cla.sifiers on Im.a.a.ced Big Da.a

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Com.uters, m.0}eria.s a.d Continua 2024年第1期81卷 1703-1728页

作者： Na.ir F. a.m.d a.a. Sa.irKira. m. Yevseyeva.I. Sa.f m. School of Com.uter Science a.d Inform.0}ics De Montfort University Leicester LE1 9BH United Kingdom Fa.ulty of Com.uter Science a.d Inform.0}ion Technology Universiti Tun Hussein Onn Malaysia Johor 86400 Malaysia

Integra.ing m.0}hine lea.ning a.d da.a.m.ning is crucia. for processing big da.a.a.d extra.ting va.ua.le insights to enha.ce decision-m.0}ing. However, im.a.a.ced ta.get va.ia.les within big da.a.present technica. cha.lenges tha. hinder the perform.0}ce of supervised lea.ning cla.sifiers on key eva.ua.ion m.trics, lim.ting their overa.l effectiveness. This study presents a.com.rehensive review of both com.on a.d recently developed Supervised Lea.ning Cla.sifiers (SLCs) a.d eva.ua.es their perform.0}ce in da.a.driven decision-m.0}ing. The eva.ua.ion uses va.ious m.trics, with a.pa.ticula. focus on the Ha.m.nic m.a. Score (F-1 score) on a. im.a.a.ced rea.-world ba.k ta.get m.0}keting da.a.et. The findings indica.e tha. grid-sea.ch ra.dom.forest a.d ra.dom.sea.ch ra.dom.forest excel in Precision a.d a.ea.under the curve, while Extrem. Gra.ient Boosting (XGBoost) outperform. other tra.itiona. cla.sifiers in term. of F-1 score. Em.loying oversa.pling m.thods to a.dress the im.a.a.ced da.a.shows significa.t perform.0}ce im.rovem.nt in XGBoost, delivering superior results a.ross a.l m.trics, pa.ticula.ly when using the Sm.TE va.ia.t known a. the BorderlineSm.TE2 technique. The study concludes severa. key fa.tors for effectively a.dressing the cha.lenges of supervised lea.ning with im.a.a.ced da.a.ets. These fa.tors include the im.orta.ce of selecting a.propria.e da.a.ets for tra.ning a.d testing, choosing the right cla.sifiers, em.loying effective techniques for processing a.d ha.dling im.a.a.ced da.a.ets, a.d identifying suita.le m.trics for perform.0}ce eva.ua.ion. a.ditiona.ly, fa.tors a.so enta.l the utilisa.ion of effective explora.ory da.a.a.a.ysis in conjunction with visua.isa.ion techniques to yield insights conducive to da.a.driven decision-m.0}ing. © 2024 The a.thors. Published by Tech Science Press.

关键词： Big da.a da.a.m.ning da.a.visua.iza.ion im.a.a.ced da.a.et la.el encoding m.0}hine lea.ning sa.pling techniques

Coopera.ive Ta.get Fencing of m.ltiple Vehicles for a.Genera. Ta.get with Connectivity Preserva.ion a.d Collision a.oida.ce

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journa. of System. Science & Com.lexity 2024年第1期37卷 136-151页

作者： Pa. Zini CHEN Ben m Depa.tm.nt of m.cha.ica. a.d a.tom.0}ion Engineering The Chinese University of Hong KongHong KongChina

In this pa.er,the a.thors study the coopera.ive ta.get-fencing problem.for n-dim.nsiona. system. a.d a.ta.get with a.genera. *** using the velocity of the vehicles,a.position feedba.k control la. is proposed to fence the genera. ta.get into the convex hull form.d by the ***,the dyna.ics of ea.h vehicle is described by a.double-integra.or *** potentia. functions a.e designed to gua.a.tee connectivity preserva.ion of the com.unica.ion network a.d collision a.oida.ce a.ong the *** proposed a.proa.h ca. dea. with a.ta.get whose tra.ectory is a.y twice continuously differentia.le function of *** effectiveness of the result is verified by a.num.rica. exa.ple.

关键词： m.lti-a.ent system. networked system. ta.get fencing