简介：ThispaperpointsoutthelimitationsofbothparametricandgeometricmodelmethodscommonlyappliedtogenerateoperationalsketchesinCAPP.Theideasofmulti-levelgraphiclibraryareintroducedandanewgenerativemethodbasedonthislibrarywhichcombinesparametricmethodwithgeometricmodelmethodispresented.Theprinciplesandcharacteristicsoftheconstructionofthelibraryaredescribedindetailandthefeaturedatastructurecorrespondingwiththelibraryisalsopresented.

简介：BecausemodelswitchingsystemisatypicalformofTakagi-Sugeno(T-S)modelwhichisanuniversalapproximatorofcontinuousnonlinearsystems,wedescribethemodelswitchingsystemasmixedlogicaldynamical(MLD)systemanduseitinmodelpredictivecontrol(MPC)inthispaper.Consideringthateachlocalmodelisonlyvalidineachlocalregion,weaddlocalconstraintstolocalmodels.Thestabilityofproposedmulti-modelpredictivecontrol(MMPC)algorithmisanalyzed,andtheperformanceofMMPCisalsodemonstratedonaninulti-multi-output(MIMO)simulatedpHneutralizationprocess.

简介：高分辨率图象熔化是图象处理的域里的一个重要焦点。一个新图象熔化模型基于实验模式分解(EMD)的典型水平被介绍。多光谱的图象的紧张色彩浸透(代表耶稣之符号)变换首先给紧张图象。此后，以ID的排列扩展，EMD建模的2DEMD被用来从高分辨率的乐队图象和紧张图象分解详细规模图象和粗糙的规模图象。最后，一幅熔化紧张图象被重建在熔化图象与高分辨率的图象的高频率和紧张图象和代表耶稣之符号反的变换结果的低频率获得。在介绍EMD原则以后，2DEMD的一个多尺度的分解和重建算法被定义，一个熔化技术计划基于EMD是先进的。全色的乐队和多光谱的乐队3,2,1Quickbird被用来估计熔化算法的质量。在在特定的排(列)上根据EMD分析为兼并选择适当内在的模式函数(IMF)以后，象素灰色珍视系列，熔化计划给一幅熔化图象，它与通常使用的熔化算法相比(小浪，代表耶稣之符号，Brovey)。图象熔化的目的包括提高图象的可见性并且改进空间分辨率并且光谱原来的图象的信息。为了估计一幅图象的质量，在熔化，信息熵和标准差被使用估计熔化图象和相关系数的空间细节以后，为测量在原来的图象和熔化图象之间的失真以偏导索引和变弯的度光谱信息。为建议熔化算法，当EMD算法被用来执行熔化经验时，更好的结果被获得。

简介：Amulti-objectivehybridgeneticbasedoptimizationalgorithmisproposedaccordingtothemulti-objectivePropertyofinverseplanning.Itisbasedonhybridadaptivegeneticalgorithmwhichcombinesthesimulatedannealing,usesadaptivecrossoverandmutation,andadoptsnichedtournamentselection.Theresultoftheteatcalculatationdemonstratesthatanexcellentconvergingspeedcanbeachievedusingthisapproach.

简介：Inthepresentstudy,27multi-drugresistantstrainsofPseudomonasaeruginosawereisolatedfromclinicalspecimensinourhospitalfromJan2005toNov2005,inwhichtheresistantgenesencodingβ-lactamaseincludingTEM,SHV,OXA,PER,VEB,GES,CARB,IMP,VIM,SPM,GIM,DHAandOprD2weretestedbyPCRamplificationandsequencedbyDNAsequencer.Itwasfoundthatthedetectionratesofbla_(VEB),bla_(GES)andbla_(CARB)genesinthese27isolatesofP.aeruginosawere11.1%,11.1%and48.1%,respectively,butalmosttheoprD2genewaslacked(92.6%).Inaddition,theresistantgenesencodingTEM,SHV,OXA,PER,IMP,VIM,SPM,GIMandDHAβ-lactamasewereallnotfound.Itwasalsodemonstratedthatthesequenceofbla_(VEB)geneappearedtobeidenticaltothatofthebla_(VEB-1)(AY536743),whilethebla_(CES)andbla_(CARB)genesshared99%identitywithbla_(GES-1)(AY219651)andbla_(CARB-3)(S46063)genes.Fromtheseobservations,itisevidentthatP.aeruginosacarryingthebla_(VES),bla_(GES)andbla_(CARB)resistantgenesisolatedinourhospitalconferstheresistancetoβ-lactams,andthelossoftheoprD2genemaybetheimportantcausetodevelopresistancetoimipeneminP.aeruginosa.

简介：

简介：Informationonricephenologyisessentialforyieldestimationandcropmanagement.Totesttheabilityofremotesensingindetectingmultiplephenologicalstages,paddyricecanopyspectrumwasmeasuredbyahand-heldradiometer.Normalizeddifferencevegetationindex(NDVI)wascalculatedfromspectrum,andtheslopeofNDVIwasobtainedasitsdifference.WeevaluatedtheresponseofNDVIanditsslopetoricegrowthwithacomparisonoftwolate-seasonricecultivars.TheresultsshowedthatNDVIanditsslopecurveshaddistinctvariationcorrespondingtoricedevelopmentandtheycouldbeusedascultivar-independentphenologicalindicators.Thedatesoffloodingandtransplanting,tillering,panicledevelopment,headingandflowering,maturity,harveststages,andevenfieldmanagementpractices,couldbededucedfromtheseindicators.‘NDVI≤0'couldbeusedasasinglethresholdforthedetectionoffloodingandtransplanting.ThelargestspikeinthecurveoftheNDVIslopeindicatedthedurationoftilleringstage.Thenextspikecorrespondedtopanicledevelopment.TheheadingandfloweringstagewascharacterizedbythemaximumNDVIandthechangeofNDVIslopefrompositivetonegative.Atthematuritystage,NDVIdecreasedcontinuously,anditsslopefluctuatedjustbelowzero.Whenricegrainswerecompletelymatureandreadyforharvest,NDVIdeclinewasaccelerated.Atharvest,NDVIslopereacheditsminimumvalue.Thedistinctionbetweenheadingandmaturitystageswasobscure,mostlikelyduetoNDVIsaturationathighbiomass.Thestudymightprovidereferencesforpaddyricephenologydeterminationthroughremotesensingimages.

简介：Aspectral-based8-inkcharacterizationmodelisdevelopedtoaccuratelypredicttherecipeforamulti-inkprinter.The8-inkcolorseparationmethodisaunionoffive3-inkandsix4-inkcombinationsbasedonthecellularYule-NielsenspectralNeugebauermodelwitharecipeselectionstrategy.Theperformancelevelsoftheforwardandbackwardmodelsareevaluatedforindividualinkcombinationsusingprintedtestingsamples.Furthermore,thespectral-basedmethodperformsbettercomparedwiththeXYZ-basedapproach.Onthebasisofthebackwardmodelperformance,anovelfastrecipeselectionstrategyisproposedandestimated.

简介：Thefocusingofaradiallypolarizedbeamwithoutannularapodizationoraphasefilterattheentrancepupiloftheobjectiveresultsinawidefocusandlowpurityofthelongitudinallypolarizedcomponent.However,thepresenceofaphysicalannularapodizationorphasefiltermakessomeapplicationsmoredifficultorevenimpossible.Weproposearadiallypolarizedandamplitude-modulatedannularmulti-Gaussianbeammode.Numericalsimulationshowsthatitcanbefocusedintoasharperfocalspotof0.125λ~2withoutadditionalapodizationsorfilters.Thebeamqualitydescribingthepurityoflongitudinallypolarizedcomponentisupto86%.

简介：Onthebasisofathoroughunderstandingofthephysicalcharacteristicsofremotesensingimage,thispaperemploysthetheoriesofwavelettransformandsignalsamplingtodevelopanewimagefusionalgorithm.ThealgorithmhasbeensuccessfullyappliedtotheimagefusionofSPOTPANandTMofGuangdongprovince,China.Theexperimentalresultsshowthataperfectimagefusioncanbebuiltupbyusingtheimageanalyticalsolutionandre-constructionintheimagefrequencydomainbasedonthephysicalcharacteristicsoftheimageformation.Themethodhasdemonstratedthattheresultsoftheimagefusiondonotchangespectralcharacteristicsoftheoriginalimage.

简介：Multipleperformancerequirementsneedtobeguaranteedinsomereal-timeapplicationssuchasmultimediadataprocessingandreal-timesignalprocessinginadditiontotimingconstraints.Unfortunately,mostconventionalschedulingalgorithmsonlytakeoneortwodimensionsofthemintoaccount.Motivatedbythisfact,thispaperinvestigatestheproblemofprovidingmultipleperformanceguaranteesincludingtimeliness,QoS,throughput,QoSfairnessandloadbalancingforasetofindependenttasksbydynamicscheduling.Webuildaschedulermodelthatcanbeusedformulti-dimensionalscheduling.Basedontheschedulermodel,weproposeaheuristicmulti-dimensionalschedulingstrategy,MDSS,consistingofthreesteps.Thefirststepcanbeofanyexistingreal-timeschedulingalgorithmthatdeterminestoacceptorrejectatask.Instep2,weputforwardanovelalgorithmMQFQtoenhancetheQoSlevelsofacceptedtasks,andtomakethesetaskshavefairQoSlevelsatthesametime.AnothernewalgorithmITLBisproposedandusedinstep3.TheITLBalgorithmiscapableofbalancingloadandimprovingthroughputofthesystem.ToevaluatetheperformanceofMDSS,weperformextensivesimulationexperimentstocompareMDSSstrategywithMDSRstrategy,DASAPandDALAPalgorithms.ExperimentalresultsshowthatMDSSsignificantlyoutperformsMDSR,DASAPandDALAP.

简介：在这篇论文，支持向量基于机器的多模型预兆的控制被建议，在哪个SVM分类与SVM结合很好回归。起初，每工作环境被SVM回归和支持向量机器建模基于网络的模型相应于每环境的预兆的控制(SVMN-MPC)算法被开发，然后一个多班SVM模型被建立认出多重操作条件。至于控制，当前的环境被多班SVM模型识别然后相应SVMN-MPC控制器在每采样片刻被激活。当模特儿建议，切换和控制器设计在模拟结果被表明。

简介：Usingbi-spinorfieldswewritethepseudo-scalarandbi-spinorfieldsthatarecharacterizedbythefieldfunctionsofcoordinatesofseveralparticles,namelymulti-particlefields.Byapplyingthequantizationproceduretothesemulti-particlefields,hadroniccreationandannihilationoperatorshavebeenobtained.

简介：Basedonthediffractiontheorymodelforhot-imageformation,theevolutionofhot-imagesinducedbymulti-scattererslocatedinthesameplaneperpendiculartothepropagationaxisisnumericallysimulated.Thesimulationresultsshowthathot-imagesinducedbycoplanarmulti-scatterersarealsocoplanarnomatterwhethertheyexistsimultaneouslyorseverally.However,iftheyexistsimultaneouslythepeakintensityoftheprimaryhot-imagesisweakerthaniftheyexistseverally.Theunequalcompetitionforenergybetweenthescatteredbeamsfromthescatterersleadstothefactthatpartofthecorrespondinghot-imagesarerelativelyenhancedandtheothersarerestrained.Theresultsshowthatthehot-imagesofcertainscatterersbecomestrongerwhenanyoftheseparameters,i.e.amplitudemodulationcoefficient,phasemodulationcoefficientandsizeofthesurroundingscatterer,decrease.

简介：在这份报纸，我们建议了能与深convolutional从食物图象识别盘子类型，食物成分，和煮的方法的一个多工系统神经网络。我们为每个班与至少500幅图象建立了不同食物的360个班的数据集。到数据的噪音，它是从因特网收集了的还原剂，孤立点图象通过与深convolutional特征训练的一个类的SVM被检测并且消除。我们同时训练了一个盘子标识符，一个煮的方法识别器，和一个多标签成分察觉者。他们在深网络体系结构分享一些低级的层。建议框架与手工制作的特征，和识别器和成分察觉者能被用于没在训练数据集被包括为用户提供引用信息的盘子的煮的方法比传统的方法显示出更高的精确性。

简介：

简介：充分提取并且我的在时空尺寸的水库和地质的结构的多尺度的特征，新3D多尺度的容量的弯曲(MSVC)方法论在这份报纸被介绍。我们也为计算3D建议一个快算法容量的弯曲。与常规容量的弯曲属性相比，它的主要改进和关键算法在wavenumber域在时间频率领域和相应多尺度的适应微分操作符介绍多频率部件扩大，进容量的弯曲计算。这方法论能同时在两时空描绘地震多尺度的特征。另外，我们在各种各样的规模使用容量的弯曲的数据熔化充分利用弯曲大小在不同规模提取的地质的特征和异例。3DMSVC能加亮地质的异例并且同时减少噪音。因此，它改进弯曲属性分析的解释效率。3DMSVC被用于土地和海洋的3D地震数据。结果证明它能显示水库的空间分发，检测差错和破裂地区，并且识别他们的多尺度的性质。

简介：我们在场与多重奇特测量光旋涡的拓扑的费用的一个简单方法。用一个圆柱的透镜，一根旋涡横梁能与多重分开的黑暗的洞腐烂进轻地分布，谁的数，就等于输入横梁的拓扑的费用。这个结论然后与多重奇特经由实验和旋涡横梁的繁殖的数字模拟被验证。这个方法也是可靠的与奇特的不同分布测量宽带旋涡横梁的拓扑的费用，它不诉诸多重横梁interferometric实验。

简介：Thispaperpresentsalarge-scaleparticletrackingvelocimetry(LSPTV)systemformeasuringsurfacevelocityinvastunsteadyflows.Thesystemconsistsofaflexiblenumberofone-computer-six-cameraworkingunitsconnectedvialocalnetworkortheinternet.Multiplezero-tiltingcamerasareusedtoincreasesthefieldofviewwithminimumperspectivedistortions.Externalsynchronizationofvideooutputsofallcamerasfacilitatesuninterruptedimageacquisitionwithinthesystem.Effectivenessandefficiencyofthemulti-channelLSPTVareenhancedbyoptimizingillumination,flowseeding,imageacquisitionandcorrection,traceridentification,particlematching,andpost-processing.ThecapacityoftheLSPTVwasdemonstratedinthemeasurementofphysicalmodelflowsfortheThreeGorgesProject.

简介：