Logo do repositório
 
Publicação

Exploring the use of classification uncertainty to improve classification accuracy

2021-06-28Documento de conferência Acesso aberto
Ver registo simples
dc.contributor.authorMoraes, Daniel
dc.contributor.authorBenevides, P.
dc.contributor.authorMoreira, F. D.
dc.contributor.authorCosta, Hugo
dc.contributor.authorCaetano, Mário
dc.contributor.institutionNOVA Information Management School (NOVA IMS)
dc.contributor.institutionInformation Management Research Center (MagIC) - NOVA Information Management School
dc.date.accessioned2021-10-04T23:19:03Z
dc.date.available2021-10-04T23:19:03Z
dc.date.issued2021-06-28
dc.descriptionMoraes, D., Benevides, P., Moreira, F. D., Costa, H., & Caetano, M. (2021). Exploring the use of classification uncertainty to improve classification accuracy. In The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLIII-B3-2021, XXIV ISPRS Congress (2021 edition), (pp. 81-86). https://doi.org/10.5194/isprs-archives-XLIII-B3-2021-81-2021
dc.description.abstractSupervised classification of remotely sensed images has been widely used to map land cover and land use. Since the performance of supervised methods depends on the quality of the training data, it is essential to develop methods to generate an enhanced training dataset. Active learning represents an alternative for such purpose as it proposes to create a dataset of optimized samples, normally collected based on classification uncertainty. However, it is heavily dependent on human interaction, since the user has to label selected samples over a number of iterations. In this paper, we explore the use of uncertainty to improve classification accuracy through a single iteration. We conducted experiments in a region of Portugal (Trás-os-Montes), using multioral Sentinel-2 images. The proposed approach consisted in computing the classification uncertainty of a Random Forest to collect additional training data from areas of high uncertainty and perform a new classification. An accuracy assessment was performed to compare the overall accuracy of the initial and new classifications. The results exhibited an increase in accuracy, though considered not statistically significant. Obstacles related to labelling additional sampling units resulted in a lack of additional training data for various classes, which might have limited the accuracy improvement. Additionally, an uneven proportion of additional training sampling units per class and the collection of new sample data from a limited number of uncertainty regions might also have prevented a higher increase in accuracy. Nevertheless, visual inspection of the maps revealed that the new classification reduced the confusion between some classes.en
dc.description.versionpublished
dc.format.extent6
dc.format.extent1562903
dc.identifier.doi10.5194/isprs-archives-XLIII-B3-2021-81-2021
dc.identifier.issn1682-1750
dc.identifier.otherPURE: 34089330
dc.identifier.otherPURE UUID: ebcdf239-5a5a-4326-b84b-e4328b5cc889
dc.identifier.otherScopus: 85115859687
dc.identifier.urihttp://hdl.handle.net/10362/125621
dc.identifier.urlhttps://www.scopus.com/pages/publications/85115859687
dc.language.isoeng
dc.peerreviewedyes
dc.relationinfo:eu-repo/grantAgreement/FCT/3599-PPCDT/PCIF%2FMOS%2F0046%2F2017/PT
dc.subjectAccuracy Assessment
dc.subjectClassification Uncertainty
dc.subjectLand Cover
dc.subjectRemote Sensing
dc.subjectSupervised Classification
dc.subjectInformation Systems
dc.subjectGeography, Planning and Development
dc.subjectSDG 15 - Life on Land
dc.titleExploring the use of classification uncertainty to improve classification accuracyen
dc.typeconference object
degois.publication.firstPage81
degois.publication.lastPage86
degois.publication.titleThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLIII-B3-2021 XXIV ISPRS Congress (2021 edition)
degois.publication.title2021 24th ISPRS Congress Commission III: Imaging Today, Foreseeing Tomorrow
dspace.entity.typePublication
oaire.awardNumberPCIF/MOS/0046/2017
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/3599-PPCDT/PCIF%2FMOS%2F0046%2F2017/PT
oaire.fundingStream3599-PPCDT
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
rcaap.rightsopenAccess
relation.isProjectOfPublication42429afc-e9ee-44e2-8443-b8741ec75623
relation.isProjectOfPublication.latestForDiscovery42429afc-e9ee-44e2-8443-b8741ec75623

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
Use_Classification_uncertainty_Improve_Classification_Accuracy.pdf
Tamanho:
1.49 MB
Formato:
Adobe Portable Document Format
Ver/Abrir

Coleções

NIMS: MagIC - Documentos de conferências internacionais