Premium
VGI and Reference Data Correspondence Based on Location‐Orientation Rotary Descriptor and Segment Matching
Author(s) -
Mohammadi Nazila,
Malek Mohammadreza
Publication year - 2015
Publication title -
transactions in gis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.721
H-Index - 63
eISSN - 1467-9671
pISSN - 1361-1682
DOI - 10.1111/tgis.12116
Subject(s) - volunteered geographic information , matching (statistics) , outlier , orientation (vector space) , computer science , data mining , pattern recognition (psychology) , feature (linguistics) , information retrieval , artificial intelligence , mathematics , statistics , data science , linguistics , philosophy , geometry
Volunteered geographic information ( VGI ) is spatial data that has been contributed by numerous volunteers in the form of user‐generated content. VGI provides the ability for non‐experts to collect and share geographic information. Comparison of a VGI dataset with a reference dataset in order to find data correspondence is a common process in some analyses of VGI data, such as quantitative analysis of its quality. This study addresses this case and presents an automated feature matching method for VGI linear data. The proposed method consists of two main steps. In the first step, a feature point matching method based on a novel descriptor, named Location‐Orientation Rotary Descriptor ( LORD ), is performed. The proposed LORD descriptor uses the location and orientation information of the linear data in a log polar structure and in a rotary manner. In the second step, a new segment matching strategy based on the buffered linear features is performed to reject the outliers of the previous step. The usefulness of the proposed method was demonstrated by applying it to the OpenStreetMap ( OSM ) dataset, as a VGI dataset, using an official reference dataset.