预览加载中，请您耐心等待几秒...

基于图论的灰度图像分割.docx

预览
1/3
2/3
3/3

在线预览结束，喜欢就下载吧，查找使用更方便

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

基于图论的灰度图像分割 Abstract Imagesegmentationisanimportanttaskincomputervisionandimageprocessing.Itistheprocessofdividinganimageintomultiplesegmentsorregionsbasedoncertaincriteria.Inrecentyears,graphtheoryhasbeenwidelyusedinimagesegmentationduetoitsabilitytorepresentanimageasagraphandhandlecomplexrelationsbetweenpixels.Thispaperpresentsareviewofgraph-basedimagesegmentationtechniques,particularlythosebasedongrey-levelimages.Wediscussthemainstepsofgraph-basedimagesegmentationmethodsandtheadvantagesanddisadvantagesofdifferentgraphmodelsandalgorithms.Wealsopresentsomerecentadvancesingraph-basedimagesegmentationandfutureresearchdirections. Introduction Imagesegmentationistheprocessofdividinganimageintomultipleregionsorsegments,whereeachsegmentrepresentsameaningfulpartoftheimage.Imagesegmentationisacriticalstepinmanycomputervisionandimageprocessingapplications,suchasobjectrecognition,tracking,andreconstruction.Imagesegmentationcanbedonebasedonmanycriteria,includingcolor,texture,edges,andregion-based.Inrecentyears,graph-basedimagesegmentationhasgainedconsiderableattentionduetoitsabilitytorepresentanimageasagraphandhandlecomplexrelationsbetweenpixels.Inthispaper,wefocusongraph-basedsegmentationtechniquesbasedongrey-levelimages. Graph-basedimagesegmentation Thegraph-basedimagesegmentationmethodscanbedividedintotwocategories,namely,globalandlocalmethods.Globalmethodsaimtominimizeaglobalenergyfunctionthatmeasuresthedissimilaritybetweensegments.Ontheotherhand,localmethodsstartwithaninitialseedpointorregionandgrowthesegmentationbyiterativelymergingorsplittingpixelsbasedonlocalcriteria.Inthissection,wediscussthegeneralstepsofgraph-basedimagesegmentationmethods. Graphrepresentation Graph-basedimagesegmentationmethodsstartbyrepresentingtheimageasagraphG=(V,E),whereeachvertexv∈Vcorrespondstoapixelintheimageandeachedgee∈Erepresentstherelationshipbetweenadjacentpixels.Theedgescanbeweightedbasedonthesimilarityordissimilaritybetweenthepixelstheyrepresent.Thegraphcanberepresentedbyanadjacencyma