Please use this identifier to cite or link to this item: http://localhost:8080/xmlui/handle/123456789/2526
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSowmiya, B-
dc.contributor.authorSaminathan, K-
dc.contributor.authorChithraDevi, M-
dc.date.accessioned2024-05-08T10:13:14Z-
dc.date.available2024-05-08T10:13:14Z-
dc.date.issued2024-05-08-
dc.identifier.issn2286-9131-
dc.identifier.urihttp://localhost:8080/xmlui/handle/123456789/2526-
dc.description.abstractPaddyisacrucialfo o dcropprovidingessentialnutrientsandenergyandservingmorethanhalftheglobalp opulation.Diagnosingandpreventingplantdiseasesatanearlystageiscrucialforthehealthandpro ductivityofcrops.Automateddiseasediagnosiseliminatestheneedforexp ertsanddeliversaccurateoutcomes.ThisresearchwilldiagnosepaddyleafdiseaseswithDeepLearningtechnology.Thediseasessuchasbacterialblight,blast,tungro,brownsp ot,andhealthyleafclassesarediagnosedandclassi edinthisstudy.Thedatasetcontains160imagesfromeachclasswith800im-ages.Ourprop osedmo delisanensembleoftransfer-learnedInceptionV3andVGG16architectures,whichutilizesthestrengthofindividualmo d-elstoimproveoverallp erformance.Theuseoftransfer-learnedensembledeeplearningarchitecturesachievedimpressiveaccuracyratesof97.03%,94.97%,and98.87%fortraining,validationandtestingresp ectively.Theresultsindicatingthatmo delisnotover tandgeneralizeswelltounseendata.Themo del'sp erformanceisevaluatedwithconfusionmatrixwiththeparameterslikeprecision,recall,F1-score,andsupp ort.Wealsotestedthemo del'sp erformanceagainstotherprop oseddeeplearningtechniqueswithandwithouttransferlearningtechniques.Moreover,thisresearchad-vancesreliableautomateddiseasedetectionsystems,fosteringsustainableagricultureandenhancingglobalfo o dsecurityen_US
dc.language.isoenen_US
dc.publisherBharathidasa Universityen_US
dc.subjectPaddyLeafDis-eases,Classi cation,TransferLearning,Ensemble,Incep-tionetV3,VGG16,DeepLearningen_US
dc.titleAnEnsembleofTransferLearningbasedInceptionV3andVGG16Mo delsforPaddyLeafDiseaseClassi cationen_US
dc.typeArticleen_US
Appears in Collections:Other Departments

Files in This Item:
File Description SizeFormat 
91 (2).pdf1.74 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.