Catalogue en ligne IGN

Modeling and visualizing dynamic landscape objects and their qualities / Daniel van de Vlag (2006)

Public

Titre :	Modeling and visualizing dynamic landscape objects and their qualities
Type de document :	Thèse/HDR
Auteurs :	Daniel van de Vlag, Auteur ; Alfred Stein, Directeur de thèse ; Menno-Jan Kraak, Directeur de thèse
Editeur :	Enschede [Pays Bas] : University of Twente
Année de publication :	2006
Collection :	ITC Dissertation num. 132
Importance :	170 p.
Format :	21 x 30 cm
ISBN/ISSN/EAN :	978-90-8504-384-3
Note générale :	bibliographie
Langues :	Français (fre)
Descripteur :	[Vedettes matières IGN] Analyse spatiale [Termes IGN] analyse multivariée [Termes IGN] arbre de décision [Termes IGN] classification ascendante hiérarchique [Termes IGN] données spatiotemporelles [Termes IGN] incertitude des données [Termes IGN] objet géographique [Termes IGN] ontologie
Index. décimale :	THESE Thèses et HDR
Résumé :	(auteur) This thesis focuses on modeling and visualizing dynamic landscape objects and their qualities. It contains ontologies to characterize and model dynamic landscape features using spatial data. It considers their spatial data qualities and visualizes them by explorative methods. In this study, the dynamic landscape features are derived from a coastal movement application within the Netherlands, whereby beaches are subject to nourishment due to severe erosion. The description and classification of beach objects and their processes essentially grounds on the perception of the coastal landscape. Modeling a landscape is a basic agreement on the conceptualization of these features and processes. The aim is to develop a framework for conceptualization of dynamic beach objects, to understand the physical processes involved and to illustrate decision rules adopted in classification of these objects. Also, quality issues related to beach nourishments are studied, visualized and explored, using new visualization techniques. A domain-specific ontology can serve as a framework for the conceptualization of beach objects and their processes. The discrimination into product and problem ontology supports the guidance for classification of these objects and to elucidate which data ‘fit for use’. Data qualities are assessed using a quality matrix, where ontological features are portrayed against quality elements. Elements of positional, thematic and temporal accuracy and data completeness are considered of high importance for the beach nourishment application. The problem and product ontology helps to define two scenarios; the first determined by the regulations from the Ministry for Public Works; the second grounded on the abilities from an existing spatial dataset. A comparison between them shows that 72.8% of the objects suitable and non-suitable for nourishment are correctly classified. A higher overlap is found in areas where actual beach nourishments were carried out. Inaccuracies in attributes, i.e. altitude, vegetation and wetness, influence the determination of the objects. A sensitivity analysis applied on altitude shows that determinate boundaries for beach nourishment objects are not reasonable and consequently should be treated as vague objects. The ontology for beach objects is extended with a spatio-temporal ontology that considers objects to be vague and dynamic. It contains full membership functions for crisp objects, partial membership functions for fuzzy objects and temporal membership functions for dynamic fuzzy objects. The temporal membership functions include seasonal changes of vegetation and daily changes in wetness. A sensitivity analysis shows that the calculated beach nourishment volumes are practically insensitive in relation to assumptions on the temporal membership functions. A spatio-temporal ontology, as an extent of a spatial ontology, is shown to model dynamic processes in landscape studies in a more realistic way. To classify a coastal landscape, I also consider the level of scale. Object hierarchy is essential but is often ignored when collecting and classifying landscape features. A fuzzy decision tree considers a hierarchical structure for classification based on decision rules on object attributes. These attributes are defined on the basis of uncertain parameters that may change in space and time. A Bayesian hierarchical model deals with modeling and handling this uncertainty. In the beach management application, Bayesian hierarchical modeling is applied to obtain posterior probability distributions for several boundary regions. The posterior distributions yield lower and upper limits of membership functions describing boundaries between object classes. In this way, a proper fuzzy decision tree is build that includes the inherent dynamic uncertainty. The spatial information of the application contains large multivariate and multi-temporal datasets. An integrated prototype for visualization and exploration of multivariate spatiotemporal datasets is introduced. It is applied to understand and explain the behaviour of dynamic beach objects and their uncertainties. It consists of the map environment (MAP), a parallel coordinate plot environment (PCP) for visualizing attributes of the dataset, and a temporal ordered space matrix environment (TOSM) for presenting spatio-temporal patterns. The TOSM is a new exploration method and can be seen as a schematized map, whereby the rows in the TOSM environment represent time, the columns represent geographic units, and individual cells are colored according to the value of user defined attributes. The prototype is applied on four case studies. A usability test is performed to test for the differences in the ability to detect patterns in multivariate spatio-temporal datasets for each environment. Test measures are efficiency, effectiveness and user’s satisfaction. Results show that the TOSM environment and the integrated prototype have significantly better performances in efficiency and user’s satisfaction than the MAP and PCP environment.
Note de contenu :	1: Introduction 2: Ameland case study 3: An application of problem and product ontologies for the revision of beach nourishments 4: Modeling Dynamic Beach Objects Using Spatio-temporal Ontologies 5: Incorporating Uncertainty via Hierarchical Classification using Fuzzy Decision Trees 6: Temporal Ordered Space Matrix: Representation of Multivariate Spatio-temporal Data 7: Conclusions
Numéro de notice :	17248
Affiliation des auteurs :	non IGN
Thématique :	GEOMATIQUE
Nature :	Thèse étrangère
Note de thèse :	PhD thesis : Géomatique : ITC : 2006
En ligne :	http://library.wur.nl/WebQuery/wurpubs/348623
Format de la ressource électronique :	URL
Permalink :	https://documentation.ensg.eu/index.php?lvl=notice_display&id=81878

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
17248-01	THESE	Livre	Centre de documentation	Thèses	Disponible

IGN

Centre de documentation
scientifique

Accueil

Sélection de la langue

Adresse

Se connecter

Actualité

L'actu ! Horaires du CDOS

Informations pratiques

Exemplaires(1)

IGN / ENSG

L'IGN a pour vocation

Accès directs

2014 -2024 IGN

IGN

Centre de documentationscientifique

Accueil

Sélection de la langue

Adresse

Se connecter

Actualité

L'actu ! Horaires du CDOS

Informations pratiques

Exemplaires(1)

IGN / ENSG

L'IGN a pour vocation

Accès directs

2014 -2024 IGN

Centre de documentation
scientifique