a novel text detectin and localization method based oncorner response

a novel text detectin and localization method based oncorner response

ID:40557811

大小:390.28 KB

页数:4页

时间:2019-08-04

a novel text detectin and localization method based oncorner response_第1页
a novel text detectin and localization method based oncorner response_第2页
a novel text detectin and localization method based oncorner response_第3页
a novel text detectin and localization method based oncorner response_第4页
资源描述:

《a novel text detectin and localization method based oncorner response》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、ANOVELTEXTDETECTIONANDLOCALIZATIONMETHODBASEDONCORNERRESPONSELiSun,GuizhongLiu,XuemingQian,DanpingGuoSchoolofElectronicsandInformationEngineering,Xi’anJiaotongUniversity,710049,ChinaABSTRACTmethodholdtheassumptionthatbackgroundsaremuchmoresmooththantextregions.SoitispossibletoclassifytextInfo

2、rmationoftextinvideosandimagesplaysanimpor-regionandnon-textregionaccordingtoedgeortextureinten-tantroleinsemanticanalysis.Inthispaper,weproposeansity.Buthowtoreducenoisescomingfromcomplexback-effectivemethodfortextdetectionandlocalizationinnoisygroundsisstillanopenproblem.Lyuetal.proposeamet

3、hodbackground.Thealgorithmisbasedoncornerresponse.fordetectingmultilingualandmulti-resolutiontext[4].SobelComparedtonon-textregions,thereoftenexistdenseedgesedgemapsareusedasfeaturesandaspeciallocalthresholdandcornersintextregions.Sowecangetrelativelystrongisadoptedtotolocatethecandidateregio

4、noftext.Lietal.responsesfromtextregionsandlowresponsesfromnon-textproposeamethodbasedoncentralmomentsofimageblockregions.Theseresponsesprovideususefulcuesfortextde-[1].Theyprovethatdifferentscaletextcanbedetectedusingtectionandlocalizationinimages.Thenusingasimpleblockthisfeature;3)Machinelea

5、rningbasedmethodusefeaturesbasedthresholdscheme,wegetcandidateregionsfortext.extractedfromtextregionandnon-textregiontotrainsupportTheseregionsarefurtherverifiedbycombiningotherfeaturesvectormachineorneutralnetworkandthentextdetectionbe-suchascolorandsizerangeofconnectedcomponent.Finally,comes

6、asupervisedclassificationproblem.In[5],aSVMTextlineislocatedaccuratelybytheprojectionofcornerre-basedalgorithmusingresponsefromstrokefilterisproposed.sponse.TheexperimentalresultsshowtheeffectivenessofHuetal.proposeanadaptiveSVMbasedparadigm,basedourmethods.onmaximumgradientdifferencesandotherc

7、onnectedcom-IndexTerms—Textdetection,textlocalization,cornerponentfeatures,whichcanobtainrelativelylowfalserate[6].responseTheshortcomingofmachinelearningbasedmethodisthatitneedsalargenumberoftrainingsamplesofdifferentkind.Inthispaper,anewtex

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。