DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Exploring high Resolution Satellite Image Collections Using their High-Level Features

Bahmanyar, Reza and Yao, Wei and Cui, Shiyong and Loffeld, Otmar and Datcu, Mihai (2014) Exploring high Resolution Satellite Image Collections Using their High-Level Features. In: Proceedings of ESA-EUSC-JRC 2014 - 9th Conference on Image Information Mining Conference: The Sentinels Era, pp. 77-80. EU. ESA-EUSC-JRC 2014, 05-07 March 2014, Bucharest, Romania. doi: 10.2788/25852. ISBN 978-92-79-36160-9. ISSN 1831-9424.

[img] PDF

Official URL: http://bookshop.europa.eu/en/proceedings-of-9th-esa-eusc-jrc-conference-on-image-information-mining-pbLBNA26548/?CatalogCategoryID=9.EKABstN84AAAEjuJAY4e5L


Large volume of detailed features of land covers, provided by High-Resolution Earth Observation (EO) images, has attracted the interests to assess the discovery of these features by Content-Based Image Retrieval systems. In this paper, we perform Latent Dirichlet Allocation (LDA) on the Bag-of-Words (BoW) representation of collections of EO images to discover their high-level features, so-called topics. To assess the discovered topics, the images are represented based on the occurrence of different topics, we name it Bag-of-Topics (BoT). Then, the BoT model is compared to the BoW model of images based on the given human-annotations of the data. In our experiments, we compare the classification accuracy resulted by BoT and BoW representations of two different EO datasets, a Synthetic Aperture Radar (SAR) dataset and a multi-spectral satellite dataset. Moreover, we provide isualizations of feature space for better perceiving the changes in the discovered information by BoT and BoW models. Experimental results demonstrate that the dimensionality of the data can be reduced by BoT representation of images; while it either causes no significant reduction in the classification accuracy or even increase the accuracy by sufficient number of topics.

Item URL in elib:https://elib.dlr.de/94223/
Document Type:Conference or Workshop Item (Poster)
Title:Exploring high Resolution Satellite Image Collections Using their High-Level Features
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Bahmanyar, Rezareza.bahmanyar (at) dlr.deUNSPECIFIED
Yao, WeiUniversity of Siegen, GermanyUNSPECIFIED
Cui, Shiyongshiyong.cui (at) dlr.deUNSPECIFIED
Loffeld, OtmarUniversity of Siegen, GermanyUNSPECIFIED
Datcu, Mihaimihai.datcu (at) dlr.deUNSPECIFIED
Journal or Publication Title:Proceedings of ESA-EUSC-JRC 2014 - 9th Conference on Image Information Mining Conference: The Sentinels Era
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
DOI :10.2788/25852
Page Range:pp. 77-80
EditorsEmailEditor's ORCID iD
Soille, PierreJoint Research Centre, European Commission, ItalyUNSPECIFIED
Marchetti, Pier GiorgioEuropean Space Agency, ItalyUNSPECIFIED
Iapaolo, MicheleEuropean Space Agency, ItalyUNSPECIFIED
Colaiacomo, LucioEuropean Union Satellite Centre, SpainUNSPECIFIED
Datcu, Mihaimihai.datcu@dlr.deUNSPECIFIED
Keywords:Content-Based Image Retrieval, Latent Dirichlet Allocation, High-level features, Earth Observation
Event Title:ESA-EUSC-JRC 2014
Event Location:Bucharest, Romania
Event Type:international Conference
Event Dates:05-07 March 2014
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Vorhaben hochauflösende Fernerkundungsverfahren (old)
Location: Oberpfaffenhofen
Institutes and Institutions:Remote Sensing Technology Institute > Photogrammetry and Image Analysis
Deposited On:07 Jan 2015 15:54
Last Modified:31 Jul 2019 19:51

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.