Centre de documentation
scientifique

Mask R-CNN-based building extraction from VHR satellite data in operational humanitarian action: An example related to Covid-19 response in Khartoum, Sudan / Dirk Tiede in Transactions in GIS, Vol 25 n° 3 (June 2021)  

Public [article]  inTransactions in GIS > Vol 25 n° 3 (June 2021) . - pp 1213-1227 Titre : Mask R-CNN-based building extraction from VHR satellite data in operational humanitarian action: An example related to Covid-19 response in Khartoum, Sudan Type de document : Article/Communication Auteurs : Dirk Tiede, Auteur ; Gina Schwendemann, Auteur ; Ahmad Alobaidi, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 1213-1227 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse d'image orientée objet [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection du bâti [Termes IGN] échantillonnage [Termes IGN] épidémie [Termes IGN] gestion de crise [Termes IGN] HRV (capteur) [Termes IGN] image à très haute résolution [Termes IGN] image Pléiades-HR [Termes IGN] itération [Termes IGN] Soudan Résumé : Auteur) Within the constraints of operational work supporting humanitarian organizations in their response to the Covid-19 pandemic, we conducted building extraction for Khartoum, Sudan. We extracted approximately 1.2 million dwellings and buildings, using a Mask R-CNN deep learning approach from a Pléiades very high-resolution satellite image with 0.5 m pixel resolution. Starting from an untrained network, we digitized a few hundred samples and iteratively increased the number of samples by validating initial classification results and adding them to the sample collection. We were able to strike a balance between the need for timely information and the accuracy of the result by combining the output from three different models, each aiming at distinctive types of buildings, in a post-processing workflow. We obtained a recall of 0.78, precision of 0.77 and F1 score of 0.78, and were able to deliver first results in only 10 days after the initial request. The procedure shows the great potential of convolutional neural network frameworks in combination with GIS routines for dwelling extraction even in an operational setting. Numéro de notice : A2021-464 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1111/tgis.12766 Date de publication en ligne : 06/05/2021 En ligne : https://doi.org/10.1111/tgis.12766 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98060 [article]

Monitoring recovery after earthquakes through the integration of remote sensing, GIS, and ground observations: the case of L’Aquila (Italy) / Diana Contreras in Cartography and Geographic Information Science, Vol 43 n° 2 (April - May 2016)  

Public [article]  inCartography and Geographic Information Science > Vol 43 n° 2 (April - May 2016) . - pp 115 - 133 Titre : Monitoring recovery after earthquakes through the integration of remote sensing, GIS, and ground observations: the case of L’Aquila (Italy) Type de document : Article/Communication Auteurs : Diana Contreras, Auteur ; Thomas Blaschke, Auteur ; Dirk Tiede, Auteur ; Marianne Jilge, Auteur Année de publication : 2016 Article en page(s) : pp 115 - 133 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] analyse de données [Termes IGN] détection de changement [Termes IGN] Italie [Termes IGN] outil d'aide à la décision [Termes IGN] séisme [Termes IGN] système d'information géographique Résumé : (Auteur) The usefulness of remote sensing (RS), geographical information systems, and ground observations for monitoring changes in urban areas has been demonstrated through many examples over the last two decades. Research has generally focused on the relief phase following a disaster, but we have instead investigated the subsequent phases involving early recovery, recovery, and development. Our aim was to determine to what extent integration of the available tools, techniques, and methods can be used to efficiently monitor the progress of recovery following an earthquake. Changes in buildings within the Italian city of L’Aquila following the 2009 earthquake were identified from Earth observation data and are used as indicators of progress in the recovery process. These changes were identified through (1) visual analysis, (2) automated change detection using a set of decision rules formulated within an object-based image analysis framework, and (3) validation based on a combination of visual and semiautomated interpretations. An accuracy assessment of the automated analysis showed a producer accuracy of 81% (error of omission: 19%) and a user accuracy of 55% (error of commission: 45%). The use of RS made it possible for the identification of changes to be spatially exhaustive, and also to increase the number of categories used for a recovery index. In addition, using RS allowed the area requiring extensive fieldwork (to monitor the progress of the recovery process) to be reduced. Numéro de notice : A2016-282 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1080/15230406.2015.1029520 En ligne : https://doi.org/10.1080/15230406.2015.1029520 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=80853 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
032-2016021	RAB	Revue	Centre de documentation	En réserve L003	Disponible

A new geospatial overlay method for the analysis and visualization of spatial change patterns using object-oriented data modeling concepts / Dirk Tiede in Cartography and Geographic Information Science, vol 41 n° 3 (June 2014)  

Public [article]  inCartography and Geographic Information Science > vol 41 n° 3 (June 2014) . - pp 227 - 234 Titre : A new geospatial overlay method for the analysis and visualization of spatial change patterns using object-oriented data modeling concepts Type de document : Article/Communication Auteurs : Dirk Tiede, Auteur Année de publication : 2014 Article en page(s) : pp 227 - 234 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Systèmes d'information géographique [Termes IGN] analyse comparative [Termes IGN] couche thématique [Termes IGN] détection d'objet [Termes IGN] détection de changement [Termes IGN] programmation adaptée à l'objet [Termes IGN] système d'information géographique [Termes IGN] topologie Résumé : (Auteur) Traditional geographic information system (GIS)-overlay routines usually build on relatively simple data models. Topology is – if at all – calculated on the fly for very specific tasks only. If, for example, a change comparison is conducted between two or more polygon layers, the result leads mostly to a complete and also very complex from–to class intersection. A lot of additional processing steps need to be performed to arrive at aggregated and meaningful results. To overcome this problem a new, automated geospatial overlay method in a topologically enabled (multi-scale) framework is presented. The implementation works with polygon and raster layers and uses a multi-scale vector/raster data model developed in the object-based image analysis software eCognition (Trimble Geospatial Imaging, Munich, Germany). Advantages are the use of the software inherent topological relationships in an object-by-object comparison, addressing some of the basic concepts of object-oriented data modeling such as classification, generalization, and aggregation. Results can easily be aggregated to a change-detection layer; change dependencies and the definition of different change classes are interactively possible through the use of a class hierarchy and its inheritance (parent–child class relationships). Implementation is exemplarily shown for a change comparison of CORINE Land Cover data sets. The result is a flexible and transferable solution which is – if parameterized once – fully automated. Numéro de notice : A2014-315 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/INFORMATIQUE Nature : Article DOI : 10.1080/15230406.2014.901900 En ligne : https://doi.org/10.1080/15230406.2014.901900 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=33218 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
032-2014031	RAB	Revue	Centre de documentation	En réserve L003	Disponible