Home Техника Разпознаване на размити образи

Разпознаване на размити образи



Концепция

Разпознаването на размити модели е важна насока в изследването на теорията на размитите множества, а невронната мрежа е обичаен метод за заразяване с данни.

Theoreticalbasis

In1965,awell-knownAmericancyberneticexpertandprofessorattheUniversityofCaliforniaLAZadeh(1965)proposedtheconceptoffuzzysets,establishedthetheoryoffuzzysets,andcreatedAtheoreticalmethodtostudytheproblemofambiguityoruncertainty.Inthepast40years,fuzzytheoryandtechnologyhavedevelopedrapidly.Applieddisciplinesbasedonfuzzysettheory,suchasfuzzyclusteranalysis,fuzzypatternrecognition,fuzzycomprehensiveevaluation,fuzzydecisionandfuzzyprediction,fuzzyplanning,fuzzycontrol,fuzzyInformationprocessinghasplayedaveryimportantroleinindustry,agriculture,medicine,military,computerscience,informationscience,managementscience,systemscience,engineeringtechnologyandotherdisciplines,andhasbroughthugeeconomicbenefits.

В ежедневието, хората често използват усет за идентифициране на графики, текст, език и т.н., в областите на метеорологичните науки, инженерните изследвания, инженерството на околната среда, медицината, криминалните разследвания, военните и т.н. никой обект не принадлежи. Това е проблемът на разпознаването на шаблони.

Patternrecognitionisabranchofartificialintelligencethathasbeenrapidlydevelopedinthepast30years.However,thereisnoprecisedefinitionofwhatisa"pattern"orwhatapatterncanberecognizedbymachines(includinghumans).Here,wecanonlyexplainvisuallythatthereasonwhypeoplecanrecognizeimages,sounds,actions,textshapes,facialexpressions,etc.,isbecausetheyallhavecertainpatternsthatreflecttheircharacteristics.Thisinterpretationisstillatautology,andthereisnoconnotationandextensionoftheinterpretationmodelatall.EvenL.Kanal,anartificialintelligenceexpert,alsobelieves:“Ifadefinitionofamodelemergesandisproventopromotethedevelopmentoftheory,itwillmarkamajoradvancementinhumanintelligence.However,thesituationisnotItaffectsthewideapplicationofpatternrecognitioninvariousfields."

Themechanismofhumanbeingsonthepatternrecognitionprocessisstillnotveryclear.Therecognitionofconcretethingsismainlyapsychologicalphenomenon,andtherecognitionofabstractthingsismainlyathinkingphenomenon.Whenaperson'sknowledgeofspecificthings,itinvolvestherelationshipbetweenthepersonandthestimuluscausedbytheobjectivethingsinthehumansenses.Whenapersonfeelsapattern,hecombinesthisfeelingwiththegeneralconceptsorclueshehasobtainedfromhisownpastexperience,andmakesinductivereasoningjudgments.Duetothevaryingdegreesofambiguityinthecharacteristicsofobjectivethings,classicalrecognitionmethodsarebecominglessandlesssuitablefortherequirementsofobjectivereality.Fuzzyrecognitionisproducedtomeetthisrequirement.

Thefuzzysetmethodoffuzzyrecognition,namelyfuzzypatternrecognition,isausefulsupplementtotraditionalpatternrecognitionmethods,namelystatisticalmethodsandsyntacticmethods.Itcanidentifyandjudgefuzzythings.Itstheoreticalbasisisfuzzymathematics..Fuzzypatternrecognitionistheintroductionoffuzzymathematicsmethodsinpatternrecognition,andtheuseoffuzzytechnologytodesignmachinerecognitionsystemscansimplifythestructureoftherecognitionsystem,simulatethethinkingprocessofthehumanbrainmoreextensivelyandmoredeeply,soastoperformmoreeffectiveonobjectivethings.Classificationandidentification.

Relatedapplications

Economics

Withtherapiddevelopmentofelectroniccomputers,cybernetics,andsystemscience,itisnecessaryforcomputerstobeabletodealwithcomplexthingslikethehumanbrain.Withtheabilitytorecognize,itisnecessarytostudyanddealwithambiguity.Innaturalsciences,fuzzypatternrecognitioniswidelyused.Forexample,in1992,theNationalScienceFoundationinitiatedaseminartodiscussthechallengesfacedbylanguagetechnology;inOctober2000,theNationalKeyLaboratoryofPatternRecognition,InstituteofAutomation,ChineseAcademyofSciences,joinedCasacoremember.-STAR(ConsortiumforSpeechTranslationAdvancedResearch).

Разпознаване на размити образи

Withthecontinuousdevelopmentofvarioussubjects,especiallyhumanities,socialsubjectsandother"softsciences",thetrendofmathematicsandquantificationhasbeguntoappearinthesefields.Fuzzypatternrecognitionisnolongersimplylimitedtotheapplicationofnaturalsciences,butalsoappliedtosocialsciences,especiallyeconomicmanagementdisciplines.

Forexample,ChenShouyu(2001)proposedafuzzyrecognitionmodelandmethodforsustainabledevelopmentsystemevaluation,includingtheprincipleandmethodoffuzzysafetydecision-makingtodeterminetheweightvectorofevaluationindicators.Themodelandmethodcanbeusedintheevaluationofsustainabledevelopmentsystemssuchassocialeconomy,ecologicalenvironment,resources,andenergy.ChenShouyu(2002)proposedafuzzypatternrecognitionmodeltodeterminethecentralcitybasedontheambiguityoftheconceptof"centralcity"ineconomiczoning.Healsoproposedafuzzydecisionanalysismethodtodeterminetheweightofmulti-objectiveindicators.Bydeterminingtherelativemembershipdegreeoftheindicatortothe"importance"ofthefuzzyconcept,theobjectiveweightisdetermined,avoidingthesubjectivityoftheweight.ZhangShoufengetal.(2003)usedtriangularfuzzynumberstorepresentfuzzyconcepts,proposedanewmulti-layermulti-levelfuzzypatternrecognitionmodel,andusedthismodeltoperformfuzzycomprehensiveevaluationandpatternrecognitiononacertainenterprise'scompetitiveness.Onthebasisoftraditionalfuzzypatternrecognition,WangYing(2004)usesnormalmembershipcloudtoreplacetheprecisemembershipfunctionintraditionalfuzzyrecognitionmethods,constructsarelatednormalcloudmodel,andmakesanapplicationofcloudtheoryinthefieldofeconomicmanagement.Preliminarydiscussionandidentificationofthecompetitivepositioningoftheenterpriseinthemarketovercomesthetheoreticaldefectoflosingambiguityduetotheuniquenessofthedegreeofmembershipdetermination,sothattheobtainedjudgmentconclusionsaremorereasonableandclosetoreality.WangZhongbinetal.(2004)usedthemethodofcombiningsystemandfuzzyanalysistofirstdecomposetheassetrisksfacedbyfinancialinstitutionsintoseveralriskfactorsubsystems;thenusefinancialtheoryasaguidetoestablishthemembershipdegreeofeachsubsystem’sriskfactorevaluationMatrix,andonthisbasis,thefuzzyindirectrecognitionmodelandtheconceptofclosenessareusedtoconstructafuzzyrecognitionmodelofsubsystemfactors;finally,thefuzzyrecognitionmodeloftheentirefinancialinstitutionsystemisestablishedthroughthesynthesisofthefuzzyrecognitionmodelsofvarioussubsystemstoachieveThepurposeofrankingtherisksoftheentirefinancialinstitution,resolvingfinancialhiddendangersandensuringnationaleconomicsecurity.

PublicFinance

Publicfinanceisaneconomicactivity(distributionactivity)withthestate(government)asthemainbody.Itisaneconomicactivityordistributionactivitythatfocusesonmeetingthepublicneedsofsociety.Socialpublicneedsdeterminetheexistenceofpublicfinances,thescopeofpublicfinancesandtheireffects.Publicfinanceshouldfocusonmeetingthepublicneedsofmembersofsociety,andshouldnotgobeyondmarketforcestomeettheprivateneedsofmembersofsociety.

Tojudgethesituationofpublicfinances,inapracticalsense,itisnecessarytocomprehensivelyconsidertwoaspects:oneistomakeacomprehensivejudgmentonthestateofpublicfinancestodeterminewhatthestateofpublicfinancesisatthistime,Inordertofacilitatetheadoptionofcorrespondingmeasures;thesecondistosorttheprosandconsofavarietyofsituationsinpublicfinancialmanagement,andtojudgewhetherthesituationisgoodorbad,soastohelpaccumulatemanagementexperienceandsupplementresearchmaterials.Thesetwotypesofissuesareatthesametimeinterrelated,andthelatterprovidesamorethoroughanddetailedbasisforthejudgmentoftheformer.

Inviewofthefuzzinessoftheevaluationofthequalityofpublicfinances,thefuzzypatternrecognitiontheorycanbeusedintheevaluationoftheadvantagesanddisadvantages,andtheevaluationindexsystemcanbedesigned.Comprehensiveevaluationandacomprehensiverankingofpublicexpendituresin10provinces.

(1)Установяване на модел

Firstofall,itisnecessarytodeterminetheindexsystemusedforjudgment,thatis,todeterminethedomainXforfuzzyrecognition.

Thisarticletakesthepublicexpenditureinmycountry’spublicfinanceastheobjectofidentification.Here,wewilldeterminethecaliberofthestudyin10provinces(Beijing,Shanghai,Zhejiang,Guangdong,Shanxi,Hubei,Hunan,Shaanxi),Yunnan,Gansu),andthefiscalexpendituresystemisvaguelydeterminedasbasicconstructionexpenditure,agricultureandforestryexpenditure,educationexpenditure,medicalandhealthexpenditure,socialsecuritysubsidyexpenditure,administrativemanagementexpenditure,publicsecurity,procuratorateandlawdepartmentexpenditure,andspecialexpenditure.TheyconstitutedomainX:

={capitalconstructionexpenditure,agricultureandforestryexpenditure,educationexpenditure,medicalandhealthexpenditure,socialsecuritysubsidyexpenditure,administrativemanagementexpenditure,publicinspectiondepartmentexpenditure,specialexpenditure}.

Accordingtothebasicprinciplesoffuzzypatternrecognition,combinedwiththelong-termaccumulationofexperienceinpublicfinancialmanagementandtheactualsituation,aseriesoffuzzysubsetsonthedomainXcanbeestablished.TheseseriesoffuzzysubsetsrepresentpublicThevariousambiguitiesoffinancialmanagement.Accordingtotheaboveassumption,letj=10,thenthefuzzysubsetA

specifiesthethresholdλaccordingtothefuzzyrecognitionmatrix.Accordingtothethresholdprinciple,thecut-inmatrixR①canbeobtained,sothattheoptimalvector.Removethisvectorandrepeatthismethodtogetthisoptimalvectorinturn.

(2)Събиране на данни

Според публичните финансови разходи на 10 провинции в моята страна (Пекин, Шанхай, Джъдзян, Гуангдонг, Шанси, Хубей, Хунан, Шанси, Юнан, Гансу) (разходи за капиталово строителство, разходи за селско и горско стопанство, разходи за образование, медицински и здравни разходи, социални разходи за субсидия за парични средства, разходи за административно управление, обществена сигурност, разходи за прокуратура и съдебно ведомство, специални разходи) като проучване на калибър.

This article is from the network, does not represent the position of this station. Please indicate the origin of reprint
TOP