Logo do repositório
 
Publicação

Improved fully convolutional network with conditional random fields for building extraction

2018-07-01Artigo científico Acesso aberto
Ver registo simples
dc.contributor.authorShrestha, Sanjeevan
dc.contributor.authorVanneschi, Leonardo
dc.contributor.institutionNOVA Information Management School (NOVA IMS)
dc.contributor.institutionInformation Management Research Center (MagIC) - NOVA Information Management School
dc.contributor.pblMolecular Diversity Preservation International (MDPI)
dc.date.accessioned2018-08-06T22:10:57Z
dc.date.available2018-08-06T22:10:57Z
dc.date.issued2018-07-01
dc.descriptionShrestha, S., & Vanneschi, L. (2018). Improved fully convolutional network with conditional random fields for building extraction. Remote Sensing, 10(7), [1135]. DOI: 10.3390/rs10071135
dc.description.abstractBuilding extraction from remotely sensed imagery plays an important role in urban planning, disaster management, navigation, updating geographic databases, and several other geospatial applications. Several published contributions dedicated to the applications of deep convolutional neural networks (DCNN) for building extraction using aerial/satellite imagery exists. However, in all these contributions, high accuracy is always obtained at the price of extremely complex and large network architectures. In this paper, we present an enhanced fully convolutional network (FCN) framework that is designed for building extraction of remotely sensed images by applying conditional random fields (CRFs). The main objective is to propose a methodology selecting a framework that balances high accuracy with low network complexity. A modern activation function, namely, the exponential linear unit (ELU), is applied to improve the performance of the fully convolutional network (FCN), thereby resulting in more accurate building prediction. To further reduce the noise (falsely classified buildings) and to sharpen the boundaries of the buildings, a post-processing conditional random fields (CRFs) is added at the end of the adopted convolutional neural network (CNN) framework. The experiments were conducted on Massachusetts building aerial imagery. The results show that our proposed framework outperformed the fully convolutional network (FCN), which is the existing baseline framework for semantic segmentation, in terms of performance measures such as the F1-score and IoU measure. Additionally, the proposed method outperformed a pre-existing classifier for building extraction using the same dataset in terms of the performance measures and network complexity.en
dc.description.versionpublishersversion
dc.description.versionpublished
dc.format.extent12632775
dc.identifier.doi10.3390/rs10071135
dc.identifier.issn2072-4292
dc.identifier.otherPURE: 5666361
dc.identifier.otherPURE UUID: d89b5f15-47a4-4397-bfa7-695f06837195
dc.identifier.otherScopus: 85050459863
dc.identifier.otherWOS: 000440332500158
dc.identifier.otherORCID: /0000-0003-4732-3328/work/151426702
dc.identifier.urihttp://www.scopus.com/inward/record.url?scp=85050459863&partnerID=8YFLogxK
dc.identifier.urlhttps://www.scopus.com/pages/publications/85050459863
dc.language.isoeng
dc.peerreviewedyes
dc.subjectBuilding extraction
dc.subjectConditional random fields
dc.subjectDeep convolutional neural network
dc.subjectFully convolutional network
dc.subjectHigh-resolution aerial imagery
dc.subjectGeneral Earth and Planetary Sciences
dc.subjectSDG 11 - Sustainable Cities and Communities
dc.titleImproved fully convolutional network with conditional random fields for building extractionen
dc.typejournal article
degois.publication.issue7
degois.publication.titleRemote Sensing
degois.publication.volume10
dspace.entity.typePublication
rcaap.rightsopenAccess

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
SShrestha_LVanneschi_Improved_2018.pdf
Tamanho:
12.05 MB
Formato:
Adobe Portable Document Format
Ver/Abrir

Coleções

NIMS: MagIC - Artigos em revista internacional com arbitragem científica (Peer-Review articles in international journals)