Veuillez utiliser cette adresse pour citer ce document : https://hdl.handle.net/10953/1726
Titre: Influence of Sample Size on Automatic Positional Accuracy Assessment Methods for Urban Areas
Auteur(s): Ariza-López, Francisco J.
Ruiz-Lendínez, Juan J.
Ureña-Cámara, Manuel A.
Résumé: In recent years, new approaches aimed to increase the automation level of positional accuracy assessment processes for spatial data have been developed. However, in such cases, an aspect as significant as sample size has not yet been addressed. In this paper, we study the influence of sample size when estimating the planimetric positional accuracy of urban databases by means of an automatic assessment using polygon-based methodology. Our study is based on a simulation process, which extracts pairs of homologous polygons from the assessed and reference data sources and applies two buffer-based methods. The parameter used for determining the different sizes (which range from 5 km up to 100 km) has been the length of the polygons’ perimeter, and for each sample size 1000 simulations were run. After completing the simulation process, the comparisons between the estimated distribution functions for each sample and population distribution function were carried out by means of the Kolmogorov–Smirnov test. Results show a significant reduction in the variability of estimations when sample size increased from 5 km to 100 km.
Mots-clés: accuracy
sample size
simulation
matching
automation
Date de publication: 2018
Editeur: MDPI
Collection(s) :DICGF-Artículos

Fichier(s) constituant ce document :
Fichier Description TailleFormat 
2018_Influence of Sample Size on Automatic Positional Accuracy Assessment Methods for Urban Areas.pdf2,96 MBAdobe PDFVoir/Ouvrir


Ce document est protégé par copyright


Ce document est autorisé sous une licence de type Licence Creative Commons
Creative Commons