简介：

简介：为旋转的一个新奇算法当旋转角度的波动的范围在这个区域以内是有限的时，不变的模板匹配被建议[−20°，20°]。匹配的候选人用低费用改进了的一个计算联盟者被选择关联算法。“并且”操作被采用减少计算费用。因此，算法过量地改进匹配的速度。模拟结果验证建议方法的效率。而且当引用图象的尺寸被修理时，这个节省时间的算法的优点作为实时图象的尺寸的增加更明显。建议方法的匹配的速度比分解的二水平的金字塔的速度快超过20倍加速方法。

简介：

简介：

简介：

简介：Thebraincontrolsvirtuallyallbodyfunctions,bothinternallyandininteractionwiththeexternalenvironment.Asthebasicbodyanatomyofallvertebrateshasabilateralsymmetry,structuresandfunctionsofvertebratebrainsarealsoorganizedaccordingtothisfundamentalanatomicalprincipletomeetallsensory,motor,andinternalrequirementsofbodycontrol.Consequently,particularpartsorfunctionsofthebodyarecontrolledbyparticularbrainstructures.Formammalswhosebrainsonlyhaveaverylimitedcapacityto

简介：Medicalimageapplicationinclinicaldiagnosisandtreatmentisbecomingmoreandmorewidely,Howtousealargenumberofimagesintheimagemanagementsystemanditisaveryimportantissuehowtoassistdoctorstoanalyzeanddiagnose.Thispaperstudiesthemedicalimageretrievalbasedonmulti-layerresamplingtemplateunderthethoughtofthewaveletdecomposition,theimageretrievalmethodconsistsoftworetrievalprocesswhichiscoarseandfineretrieval.Coarseretrievalprocessisthemedicalimageretrievalprocessbasedontheimagecontourfeatures.Fineretrievalprocessisthemedicalimageretrievalprocessbasedonmulti-layerresamplingtemplate,amulti-layersamplingoperatorisemployedtoextractimageresamplingimageseachlayer,thentheseresamplingimagesareretrievedstepbysteptofinishtheprocessfromcoarsetofineretrieval.

简介：ThisletterpresentsanewchunkingmethodbasedonMaximumEntropy(ME)modelwithN-foldtemplatecorrectionmodel.Firsttwotypesofmachinelearningmodelsaredescribed.Basedontheanalysisofthetwomodels,thenthechunkingmodelwhichcombinestheprofitsofconditionalprobabilitymodelandrulebasedmodelisproposed.Theselectionoffeaturesandruletemplatesinthechunkingmodelisdiscussed.ExperimentalresultsfortheCoNLL-2000corpusshowthatthisapproachachievesimpressiveaccuracyintermsoftheF-score:92.93%.ComparedwiththeMEmodelandMEMarkovmodel,thenewchunkingmodelachievesbetterperformance.

简介：

简介：在这个工作，有大约100nm和一段几测微计的一致直径的TiO2nanorods被大音阶的第五音胶化模板方法成功地准备。另外，准备的beeninvestigated.The取样的TiO2的形态学和nanorods有的结构上的先锋的臼齿的比率的影响被Fourier变换描绘红外线的光谱学(FTIR)，X光检查衍射(XRD)和扫描电子显微镜学(SEM)。XRD结果显示TiO2nanorods在锐钛矿和金红石阶段被使结晶，在退火到直到2h的400-700°C以后。

简介：Plasmoninducedhotelectronshaveattractedagreatdealofinterestasanovelrouteforphotodetectionandlightenergyharvesting.Herein,wereportahotelectronphotodetectorinwhichalargearrayofnanoconesdepositedsequentiallywithaluminum,titaniumdioxide,andgoldfilmscanbeintegratedfunctionallywithnanophotonicsandmicroelectronics.Thedeviceexhibitsastrongphotoelectricresponseataround620nmwitharesponsivityof180μA/Wundershort-circuitconditionswithasignificantincreaseunder1Vreversebiasto360μA/W.Theincreaseinresponsivityandaredshiftinthepeakvaluewithincreasingbiasvoltageindicatethatthebiascausesanincreaseinthehotelectrontunnelingeffect.Ourapproachwillbeadvantageousfortheimplementationoftheproposedarchitectureonavastvarietyofintegratedoptoelectronicdevices.

简介：

简介：Inthispaper,weproposeanovelautomaticobjectextractionalgorithm,namedtheTemplateGuidedLiveWire,basedonthepopularlyusedlivewiretechniques.Wediscussindetailsthenovelmethod'sapplicationsontongueextractionindigitalimages.Withtheguidesofagiventemplatecurvewhichapproximatesthetongue'sshape,ourmethodcanfinishtheextractionoftonguewithoutanyhumanintervention.Inthepaper,wealsodiscussedindetailshowthetemplateguidesthelivewire,andwhyourmethodfunctionsmoreeffectivelythanotherboundarybasedsegmentationmethodsespeciallythesnakealgorithm.Experimentalresultsonsometongueimagesareaswellprovidedtoshowourmethod'sbetteraccuracyandrobustnessthanthesnakealgorithm.

简介：Zr-Incorporatedmesoporoustitaniamaterialswerepreparedvianonsurfactanttemplatedsol-gelprocessofzirconium(IV)butoxide(ZBT)andtitanium(IV)butoxide(TBT)inthepresenceofureamoleculesastemplateorporeformingagent.TheeffcetsoftemplatecontentsontheporeparametersofthematerialssynthesizedwithfixedZrincorporationcontentswereinvestigatedbymitrogenadsorption-desrptionmeasurements,powderX-raydiffraction(XRD)studyandtransmissionelectronmicroscopy(TEM).ThechangesoftemplatecontentsplaysignificantrolesontheporeparametersatlowincorporsationcontentofZr.Theporediametersdisplayaclearincreasetendencywiththeincreaseoftemplatecontents.WhenhighZrincorporationcontentisused,thetemplatecontentshavelesseffectontheporediametersthatalmostkeepunchangedwiththeincreaseoftemplatecontents.AllthematerialspossesstypeIVisothermswithH2hysteresisloopssuggestingtheformationofmesophase.ThematerialswithlowZrincorporationcontenthaveanatasestructures；however,itcannotbefoundinthematerialswithhighZrincorporationcontent.TEMimagesshowthatsomeaccumulatedinter-particulateporesandwelldistributedworm-likeporesarepresentintheZr-incorporatedmaterials.

简介：K-phillipsite用一个热水的方法被准备。可溶的玻璃和钠铝酸盐当一个器官的模板不在时被用作原料。关于K+离子的调查在房间温度被进行在海水样品和混合K+Na+答案的选择系数决定离子交换能力。样品被X光检查衍射(XRD)描绘，扫描电子显微镜学(SEM)，和精力分散光谱学(版本)。K+离子交换能力是在海水的51mg/g，选择系数在混合K+Na+答案是75.1。样品为K+有选择偏爱，并且能因此被用来有选择地从海水提取钾。由Si，艾尔，K，Na，和O组成的样品展出跨like形状并且是典型K-phillipsite结构。

简介：Anewmethodforfabricatingorderedporoussiliconisreported.Atwo-dimensionalsilicananospherearrayisusedasatemplatewithahydrofluoricacid-hydrogenperoxidesolutionforetchingthenanospheres.Theinitialdiameteranddistributionoftheholesintheresultingporoussiliconlayeraredeterminedbythesizeanddistributionofthesilicananospheres.Thecorrosiontimecanbeusedtocontrolthedepthsoftheholes.ItisfoundthatthepresenceofaSiO_2layer,formedbytheoxidationoftheroughinternalsurfaceofthehole,istheprimaryreasonallowingthecorrosiontoproceed.Ultravioletreflectionandthermalconductivitymeasurementsshowthatthediameteranddistributionoftheholeshaveagreatinfluenceonpropertiesoftheporoussilicon.

简介：

简介：

简介：Prematureventricularcontraction(PVC)isthemostfrequentarrhythmiaencounteredinclinicalpractice.PVCmayoccurinhealthsubjects,whichisnotimminentlylife-threateningbutmayrequiretherapiestopreventfurtherproblems.So,thetimelyPVCrecognitionbecomesveryimportantfortheanalysisofelectrocardiogram(ECG),especiallyfortheremoteECGmonitoringusingmobilephones.Inthispaper,aconstructionmethodofpersonalizedECGtemplateandaPVCrecognitionmethodbasedontemplatematchingwerestudied.Firstly,weselected43ECGrecordingsfromtheMIT-BIHarrhythmiadatabase.Allrecordingsweredividedintotwodatasets(DS1fortrainingandDS2fortesting)andeachdatasetapproximatelycontainedthesameproportionofPVCbeats.Subsequently,foreachrecording(30min)inDS1,thefirst5minrecordingswereusedtoconstructthepersonalizedECGtemplateandthelast25minrecordingswereusedfortheR-wavepeaksdetectionandPVCrecognition,wherethetemplatematchingmethodwereused.ThevalidityoftheproposedmethodswastestedusingDS2.Theresultsshowedthat:1)highbeatdetectionaccuracywasachievedforbothPVCbeatsandnon-PVCbeats;2)thesensitivityandspecificityofPVCrecognitionwere99.11%and99.96%forthefirst5minrecordingsrespectively,99.17%and99.43%forthelast25minrecordingsrespectively.Alltheproposedmethodscanbereal-timeperformed,whichshowapromisingprospectfortheapplicationofECGmobilephones.

简介：