90
Views
6
CrossRef citations to date
0
Altmetric
Articles

Regression modeling of reduction in spatial accuracy and detail for multiple geometric line simplification procedures

ORCID Icon & ORCID Icon
Pages 47-70 | Received 26 Mar 2018, Accepted 03 May 2019, Published online: 29 Jul 2019
 

ABSTRACT

One of the important stages of map generalization is the selection of optimal simplification procedure for each spatial feature or feature type. Selected algorithms are then applied collaboratively to simplify the whole set of features. However, there is a lack of investigations that report a systematic approach of deriving a similar reduction in accuracy and detail by using different algorithms. In current paper we propose the solution to this problem on the basis of regression modeling between tolerance value of each algorithm and the value of some geometric measure which describes changes in accuracy and detail of the line. This allows fitting the regression model between tolerance values of the two selected algorithms which can be used to obtain similar simplification results. Regressions between Douglas-Peucker, Li-Openshaw and Visvalingam-Whyatt algorithm tolerance values are investigated. Application of methodology is illustrated on the example of three coastlines with significantly different spatial character. Results of the study show that regression coefficients depend highly both on the combination of the two algorithms, and on the character of the line. Finally, it is shown that a weighted combination of accuracy and detail regression models can be used to model the changes in level of detail of the line.

RÉSUMÉ

Une des étapes importante de la généralisation est le choix d’un processus optimal de simplification pour chaque objet spatial ou chaque type d’objet. Les algorithmes sélectionnés sont alors appliqués de façon collaborative pour simplifier l’ensemble des objets. Pourtant, peu de recherches portent sur une approche systématique pour obtenir un changement de précision et de niveau de détail identique en utilisant différents algorithmes. Dans ce papier nous proposons une solution à ce problème en se basant sur une modélisation par régression entre la valeur de tolérance de chaque algorithme et la valeur d’une mesure géométrique décrivant les changements en précision et niveau de détail d’une ligne. Cela permet d’ajuster le modèle de régression entre les valeurs de tolérance des deux algorithmes sélectionnés qui peuvent être utilisés pour obtenir des résultats de simplification similaires. Nous étudions les régressions entre les valeurs paramétriques des algorithmes de Douglas-Peucker, Li-Openshaw et Visvalingam-Whyatt. La méthode est appliquée sur trois lignes de côte qui ont des caractéristiques spatiales différentes. Les résultats de notre étude montrent que les coefficients de régression dépendent fortement du choix des algorithmes comparés et des caractéristiques de la ligne. Enfin, il est montré qu’une combinaison pondérée de la précision et du niveau de détail des modèles de régression peut être utilisée pour modéliser le changement de niveau de détail de la ligne.

Acknowledgements

The authors would like to thank the anonymous reviewers for their valuable comments that helped to improve the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Timofey Samsonov is a leading researcher at Lomonosov Moscow State University (MSU), Faculty of Geography, Moscow, Russia. He works in the field of automated cartography with a particular interest in algorithms for cartographic generalization and visualization of spatial data.

Olga Yakimova is an Associate Professor at Demidov Yaroslavl State University, Faculty of Mathematics, Yaroslavl, Russia. Her research focuses on computational algorithms, spatial data processing and graph theory.

Additional information

Funding

The reported study was funded by Russian Foundation for Basic Research (RFBR) according to the research project 18-07-01459-a.

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
USD 61.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
USD 487.00 Add to cart

* Local tax will be added as applicable

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.