25926.復(fù)雜背景下植物葉片圖像分割算法及其應(yīng)用研究

1、曲阜師范大學(xué)碩士學(xué)位論文復(fù)雜背景下植物葉片圖像分割算法及其應(yīng)用研究姓名：滿(mǎn)慶奎申請(qǐng)學(xué)位級(jí)別：碩士專(zhuān)業(yè)：控制理論與控制工程指導(dǎo)教師：鄭春厚20090401AbstractPlantsareimportantmeansoflivelihoodandproductionofhumanbeingsItisimportantandeSSentialforpeopletecognizeandclassifythematallefficientand

2、fastmal3_nerAtpresent，manvresearchershavepaidattentiontoplantclassificationusingtechnologyofcompmerimageDrocessingandpattemrecognitionObjectsegmentationandextractionofdigitalimageisamainbranchcausingmoreattentionindigita

3、limageprocessingandcomputervisionfieldForreasonsthatmailvalgorithmsCanbeusedtosegmentandextractobjectareacombinedwithitswidelyuseincomputervision，patternrecognition，itcausesmoreandmoreresearchers’attentionImagesegmentati

4、onisthekeystepfromimageprocessingtoimageanalysis，alsofoundationforfurtherimageanalysisOnlyaftertheproperimagesegmentation，imageanalysiscarlbemadepossibleConcerningimagesofplantsmayhavedifferentbackground，someissimple，som

5、eiscomplicate，westudieddifferentimagesegmentationalgorithmsfordifferentkindsofimagethatwithdifferentbackgroundUsingtheproperimagesegmentationalgorithm，wecangetregionofinterestfROI)inplantimage，fromwhichweCallextractshape

6、，color，ortexturefeaturesforclassifyingplantInthisthesis，wedesearchonimagesegmentationalgorithmsondifferentkindofleafimages，especiallythestudyandapplicationofimagesegmentationalgorithmsonleafimagewithcomplicatebackgroundA

7、fterimagesegmentation，weextractedshape，color，texturefeatureforclassifyingplantInthisthesis，weestablishedaplantleafimagerecognitionsystemusingplantleavestecognizeplantOurresearchworksaresummarizedasfollows：1Imagesegmentat

8、ion：Inthispaper，wedesearchonsomeclassicimagesegmentationalgorithms，suchasthresholdsegmentation，edgedetect；consideringleafimagewithcomplicatebackground，westudiedSnakes，Watershedalgorithm，andusedthesealgorithmsforsegmentin

9、gleafimages2Feanlreextraction：Weextractedshape，color，texturefeaturesofregionofinterest(ROI)intheleafimageafterimagesegmentationWedefinedformulationsofthesefeatureandhowtogetthesefeaturesarealsointroducedindetailAfterthed

10、efinitionoffeature，weextract15shapefeatures，3colorfeaturesand4texturefeaturesforclassifyingplant3Classifyingandrecognizingofplantimage：Consideringthenumberofpattemsandthedimensionalityofpatternspacearebothverylarge，westu

11、diedsomeclassifierswhicharecommonusedinimageengineering，suchas：NearestNeighborclassifier，NearestFeatureLineclassifier，andBackPropagationneuralnetworkandSupportvectormachineConsideringtmagefeaturedata，Scharacterweusesuppo