Dokumenttyp: Artikel/Aufsatz
Titel: Learning from vector data: enhancing vector-based shape encoding and shape classification for map generalization purposes
Autor*in: Knura, Martin Michael 
Erscheinungsdatum: 2024
Freie Schlagwörter: Deep learning; map generalization; vector data; feature descriptor; shape classification; building generalization
Zusammenfassung: 
Map generalization is a complex task that requires a high level of spatial cognition, and deep learning techniques have shown in numerous research fields that they could match or even outplay human cognition when knowledge is implicitly in the data. First experiments that apply deep learning techniques to map generalization tasks thereby adapt models from image processing, creating input data by rasterizing spatial vector data. Because image-based learning has major shortcomings for map generalization, this article investigates possibilities to learn directly from vector data, utilizing vector-based encoding schemes. First, we enhance preprocessing methods to match essential properties of deep learning models – namely regularity and feature description – and evaluate the performance of Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Graph Convolutional Neural Networks (GCNN) in combination with a feature-based encoding scheme. The results show that feature descriptors improve the accuracy of all three neural networks, and that the overall performances of the models are quite similar for both polygon and polyline shape classification tasks. In a second step, we implement an exemplary building generalization workflow based on shape classification and template matching, and discuss the generalization results based on a case study.
Sachgruppe (DDC): 004: Informatik
HCU-Fachgebiet / Studiengang: Geodäsie und Geoinformatik 
Zeitschrift oder Schriftenreihe: Cartography and Geographic Information Science 
Band: 51
Ausgabe: 1
Seite von: 146
Seite bis: 167
Verlag: Taylor & Francis
ISSN: 1523-0406
Verlagslink (DOI): 10.1080/15230406.2023.2273397
URN (Zitierlink): urn:nbn:de:gbv:1373-repos-12038
Direktlink: https://repos.hcu-hamburg.de/handle/hcu/934
Sprache: Englisch
Creative-Commons-Lizenz: https://creativecommons.org/licenses/by-nc-nd/4.0/
Enthalten in der SammlungPublikationen (mit Volltext)

Internformat

Seitenansichten

272
checked on 15.04.2024

Download(s)

52
checked on 15.04.2024

Google ScholarTM

Prüfe

Export

Diese Ressource wurde unter folgender Copyright-Bestimmung veröffentlicht: Lizenz von Creative Commons Creative Commons