预览加载中，请您耐心等待几秒...

基于BP神经网络的车型识别-外文翻译.doc

预览
1/10
2/10
3/10
4/10
5/10
6/10
7/10
8/10
9/10
10/10

亲，该文档总共19页，到这已经超出免费预览范围，如果喜欢就直接下载吧～

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

一、外文资料LicensePlateRecognitionBasedOnPriorKnowledgeAbstract-InthispaperanewalgorithmbasedonimprovedBP(backpropagation)neuralnetworkforChinesevehiclelicenseplaterecognition(LPR)isdescribed.Theproposedapproachprovidesasolutionforthevehiclelicenseplates(VLP)whichweredegradedseverely.Whatitremarkablydiffersfromthetraditionalmethodsistheapplicationofpriorknowledgeoflicenseplatetotheprocedureoflocationsegmentationandrecognition.Colorcollocationisusedtolocatethelicenseplateintheimage.DimensionsofeachcharacterareconstantwhichisusedtosegmentthecharacterofVLPs.TheLayoutoftheChineseVLPisanimportantfeaturewhichisusedtoconstructaclassifierforrecognizing.Theexperimentalresultsshowthattheimprovedalgorithmiseffectiveundertheconditionthatthelicenseplatesweredegradedseverely.IndexTerms-Licenseplaterecognitionpriorknowledgevehiclelicenseplatesneuralnetwork.I.INTRODUCTIONVehicleLicense-Plate(VLP)recognitionisaveryinterestingbutdifficultproblem.Itisimportantinanumberofapplicationssuchasweight-and-speed-limitredtrafficinfringementroadsurveysandparksecurity[1].VLPrecognitionsystemconsistsoftheplatelocationthecharacterssegmentationandthecharactersrecognition.Thesetasksbecomemoresophisticatedwhendealingwithplateimagestakeninvariousinclinedanglesorundervariouslightingweatherconditionandcleanlinessoftheplate.Becausethisproblemisusuallyusedinreal-timesystemsitrequiresnotonlyaccuracybutalsofastprocessing.MostexistingVLPrecognitionmethods[2][3][4][5]reducethecomplexityandincreasetherecognitionratebyusingsomespecificfeaturesoflocalVLPsandestablishingsomeconstrainsonthepositiondistancefromthecameratovehiclesandtheinclinedangles.Inadditionneuralnetworkwasusedtoincreasetherecognitionrate[6]