gbfcm自然场景图像分割方法研究-research on gb fcm natural scene image segmentation method.docx

下载文档 降价啦

15
0
约3.4万字
约 51页
2018-06-04 发布于上海
举报
版权申诉
保障服务

gbfcm自然场景图像分割方法研究-research on gb fcm natural scene image segmentation method.docx

1、本文档共51页，可阅读全部内容。
2、有哪些信誉好的足球投注网站（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。

gbfcm自然场景图像分割方法研究-research on gb fcm natural scene image segmentation method

AbstractWiththerapiddevelopmentofimagingequipment,thenaturalsceneimagesareincreasingandrelevanttopeopleslives.Becauseofthenaturalsceneimageshavethecharacteristicoflightuncertainandcomplexbackground,thesegmentationofnaturalsceneimageisdifficultandbecomeahotspotinareaofimagesegmentation.Thisdissertationtakesnaturalimagesegmentationastheobjectoftheresearch,researchesintothepopularimagesegmentationbasedongraphtheoryandFuzzyc-meansclusteringalgorithm.Themainworkisasfollows:Thepurposeofnaturalsceneimagesegmentationareintroduced,andtheproblemofimagesegmentationwhichneedtobesolvedareanalyzed.Introducesthebasictheoryofimagesegmentationandtheevaluationcriteriawidelyusedinimagesegmentation.Finally,theexperimentalresultsshowthatthisalgorithmiseffectiveandpractical.Theimagesegmentationalgorithmbasedongraphtheoryareresearched,focusesonthestudyofGraph-Based imagesegmentationalgorithm. FortheGraph-Based algorithmwhich existsundercombinedphenomenon,aweightedfunctionbasedonHSIcolorspaceandspatialinformationisproposed,andtheimprovedunion-findsetsisusedtomergetheregions.Fuzzyc-meansclusteringalgorithmisanalyzed,andanalgorithmofthenaturalsceneimagesegmentationcombinegraph-basedwithfuzzyC-meansclustering(GB-FCM)isproposed.ForfuzzyC-meansclusteringalgorithmlackofspatialinformationandlargeamountofcalculation,thegraphisdividedintosubgraphbygraph-basedalgorithm.TheamountofclustersarereducedbycalculatetheclustercentersofsubgraphwhichcanacceleratethefuzzyC-meansclustering.Forthesetofclusteringclassificationnumberisnotoptimalnecessarily,theoversegmentationmaybeexist,thenaweighteddistancebasedonthecolorofclusteringcenterandspaceinformationis propose as themergerules of union-findsets.Theimprovedunion-findsets isusedtomergetheregionsofclusteringresults.ExperimentalresultsshowthatthealgorithmbasedongraphandFCMisfeasibility.Finally,thefulltextissummarizedandprospected．IIIKeywords:imagesegmentation;Fuzzyc-meansclustering;graphtheory;union-findsets;HSIcolorspaceIV目录摘要.................................