line fitting

line fitting

ID:40351780

大小:4.66 MB

页数:27页

时间:2019-07-31

line fitting_第1页
line fitting_第2页
line fitting_第3页
line fitting_第4页
line fitting_第5页
资源描述:

《line fitting》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、LinefittingJackEltonBresenhamb.1937IBM/WinthropUniversityBresenham'sLineAlgorithm(1962)LineTracingInput:(x0,y0)&(x1,y1)Output:ApproximationofCoordinatesonthatlineLineTracing(x,y)(x+dx,y+dy)LineTracing-Pseudocode(x,y)drawLineP(x,y,x+dx,y+dy)//assumedx,

2、dy>0a=0,b=0whilea<=dxandb<=dydrawPixel(x+a,y+b)ifa/dx<=b/dya++elseb++Worksforpositivedxanddy(x+dx,y+dy)AutonomousMobileRobotsLineExtractionSplitandmergeLinearregressionRANSACHough-TransformZürichAutonomousSystemsLab4b-Perception-Features4b7LineExtract

3、ion:MotivationMapoftheASLhallwaybuiltusinglinesegments©R.Siegwart,D.Scaramuzza,ETHZurich-ASL4b-Perception-Features4b8LineExtraction:MotivationMapoftheASLhallwaybuiltusingorthogonalplanesconstructedfromlinesegments©R.Siegwart,D.Scaramuzza,ETHZurich-A

4、SL4b-Perception-Features4b9LineExtraction:MotivationWhylaserscanner:DenseandaccuraterangemeasurementsHighsamplingrate,highangularresolutionGoodrangedistanceandresolution.Whylinesegment:ThesimplestgeometricprimitiveCompact,requireslessstoragePr

5、ovidesrichandaccurateinformationRepresentsmostoffice-likeenvironment.©R.Siegwart,D.Scaramuzza,ETHZurich-ASL4b-Perception-Features4b10LineExtraction:TheProblemScanpointinpolarform:(ρi,θi)Assumptions:Gaussiannoisewith(0,σ)forρNegligibleangularuncer

6、taintyLinemodelinpolarform:xcosα+ysinα=r-π<α<=πr>=0rα©R.Siegwart,D.Scaramuzza,ETHZurich-ASL4b-Perception-Features4b11LineExtraction:TheProblem(2)Threemainproblems:Howmanylines?Whichpointsbelongtowhichline?•ThisproblemiscalledSEGMENTATIONGivenp

7、ointsthatbelongtoaline,howtoestimatethelineparameters?•ThisproblemiscalledLINEFITTINGTheAlgorithmswewillsee:1.Splitandmerge2.Linearregression3.RANSAC4.Hough-Transform©R.Siegwart,D.Scaramuzza,ETHZurich-ASL4b-Perception-Features4b12Algorithm1:Split-and

8、-Merge(standard)Themostpopularalgorithmwhichisoriginatedfromcomputervision.Arecursiveprocedureoffittingandsplitting.Aslightlydifferentversion,calledIterative-End-Point-Fit,simplyconnectstheendpointsforlinefitting.©R.Siegwart,D.Scaramuzza,ET

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。