彩色图象分割及其在中医舌图象处理中地应用.pdf

下载文档 降价啦

4
0
约5.05万字
约 53页
2016-01-10 发布于安徽
举报
版权申诉
保障服务

彩色图象分割及其在中医舌图象处理中地应用.pdf

1、本文档共53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)。

西北 1业人学倾 {_沦文:彩色f}-i象分刮,}s1:r1中医斤侧嚎处限中的i-tIB ColorImageSegmentationandItsApplicationto TongueImageProcessingforTCM Author: RenJijun Supervisor:Prof.HeMingyi Subject: BiomedicalEngineering Abstract Withthedevelopmentofmedicalsciencetechnology,basicdiagnosismethodsof TraditionalChineseMedicine(TCM)facedemandofmodernization.Viewing diagnosis,especiallythetongueisimportantcontentofbasicdiagnosismethodsof TCNt.Itcangetthephysiologicalfunctionandpathologyoftheorganismchangesby meansofobservingthetongueandhaveveryimportantmeaningtomakeadefinite diagnosisinearlierstageofalotofdiseases.However,traditionaltonguediagnosisonly dependsondoctorsobservation.Itisdifficulttomakeexactdiagnosisbecauseofthe factorofpersonality.Therefore,combiningthemodeminformationtechnologyandthe clinicexperienceofTCM experttorealizetheimpersonalityandstandardizationof tonguediagnosisofTCNIistheonlywayofTCMdevelopment. Inthetonguediagnosis,thetongueimageofpatientcanbeobtainedbyusing digitalinstrument,forexample,digitalcamera.Automatic interestedregion segmentationmustbecarriedoutbeforethetongueimageprocessing.Itmakesthe segmentationoftongueimageanimportantpartbetweenimageacquiringandimage processing.Qualityofthesegmentationwillaffectthesubsequentworkoftongue analysisdirectly.Accordingtothecharacteristicsoftongueimage,RegionOfInterest (ROI)isusedinpretreatmentoftongueimagesegmentationinthispaper.Anew algorithmthatcanautomaticallydetecttheROIwiththemethodofimageprojectionis proposed.Afteranalyzingmanyexistedalgorithmsofimagesegmentationand combiningthecharacteristicsoftongueimage.anewalgorithm thatcanchoose thresholdautomaticallyispresentedbasedonhistograminRGI3spaceofcolorimage. Becausetheprocessingincarriedoutingraylevel.thecomplexityanddatareduces notably.Inaddition,amethodbasedonself-organizing比aturemapneuralnemot长 fortheregionsegmenta