Vehicle License Plate Detection and Characters Recognition

Vehicle License Plate Detection and Characters Recognition

ID:37046437

大小:3.11 MB

页数:62页

时间:2019-05-15

Vehicle License Plate Detection and Characters Recognition_第1页
Vehicle License Plate Detection and Characters Recognition_第2页
Vehicle License Plate Detection and Characters Recognition_第3页
Vehicle License Plate Detection and Characters Recognition_第4页
Vehicle License Plate Detection and Characters Recognition_第5页
资源描述:

《Vehicle License Plate Detection and Characters Recognition》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、硕士留学生学位论文VehicleLicensePlateDetectionandCharactersRecognition作者姓名MUHAMMADSALMANInformationandCommunication学科专业Engineering指导教师Prof.YangJunMeiElectronicandInformation所在学院Engineering论文提交日期2018–06-21VehicleLicensePlateDetectionandCharactersRecognitionADissertationSubmittedfor

2、theDegreeofMasterCandidate:MuhammadSalmanSupervisor:Prof.YangJunMeiSouthChinaUniversityofTechnologyGuangzhou,China分类号:TP393学校代号:10561学号:201622856334华南理工大学硕士学位论文VehicleLicensePlateDetectionandCharactersRecognition作者姓名:MUHAMMADSALMAN指导教师姓名、:Prof.YangJunMei申请学位级别:学科专业名称:Info

3、rmationandCommunicationEngineering研究方向:论文提交日期:2018年05月27日论文答辩日期:2018年06月02日学位授予单位:华南理工大学学位授予日期:年月日答辩委员会成员:主席:委员:ABSTRACTAutomaticlicenseplaterecognition(ALPR)isimageprocessingtechnologyandhotresearchareainintelligenttransportationsystemanditshowsasignificantroleinmanyappl

4、ications.InALPRsystemlicenseplatedetectionfromtheoriginalimageisthemostimportantphase.Itisverychallengingbecauseofthevarietyofplatelayoutsandthenon-uniformoutsidelightingconditionsduringimageacquisition.Imageprocessingtechniqueiswidelyusedforlicenseplatedetection.Averityo

5、ftechniquesisproposedintheliteraturebuttheyarehavingaccuracyproblems.Inthiswork,wedevelopalicenseplatedetectionmethodusingaKNN(K-nearestneighbor)classifierwithcontour.Thesystemfindsallpossiblecontourintheimageandperformsroughcharactersearchingatdifferentscalesandanalysest

6、hecontourwidth,height,andareatotakearoughideawhetherthecontourcanbeacharacterornot.Theresultsofourapproachshowthatthismethodiscomparativelyimperviousto:1)variationsinlighting,2)licenseplateshapes,3)cameraperspective,4)shadowimagesand5)backgroundvariations.Thesystemisteste

7、dwithon150realtimeimages,capturedincampuspremisesandimagesdownloadedfrominternetunderdifferentilluminationconditionsandwegot96%results.Afterthelocalizationoflicenseplates,thealignmentofplatesiscarriedout.Basically,thisalignmenttechniquefindstheneighboroftheboundingboxofth

8、edetectedareaandfindouttheidealedgesitewheretheouterregionisquitedifferentfromtheinnerregionofth

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。