A Regression Model of Spatial Accuracy Prediction for Openstreetmap Buildings

Abstract

Abstract. Data quality assessment of OpenStreetMap (OSM) data can be carried out by comparing them with a reference spatial data (e.g authoritative data). However, in case of a lack of reference data, the spatial accuracy is unknown. The aim of this work is therefore to propose a framework to infer relative spatial accuracy of OSM data by using machine learning methods. Our approach is based on the hypothesis that there is a relationship between extrinsic and intrinsic quality measures. Thus, starting from a multi-criteria data matching, the process seeks to establish a statistical relationship between measures of extrinsic quality of OSM (i.e. obtained by comparison with reference spatial data) and the measures of intrinsic quality of OSM (i.e. OSM features themselves) in order to estimate extrinsic quality on an unevaluated OSM dataset. The approach was applied on OSM buildings. On our dataset, the resulting regression model predicts the values on the extrinsic quality indicators with 30% less variance than an uninformed predictor.

A Classification Model for the Inference of Spatial Precision of Openstreetmap Buildings With Intrinsic Indicators

“…In (Hashemi and Ali Abbaspour, 2015), the authors use the editing history to identify topological inconsistencies. Among approaches using only the data themselves, some use spatial context to analyze the spatial consistency of data (see (Touya and Brando-Escobar, 2013), where the author specifically identify level of detail inconsistencies), while others rely on the geometries of individual objects (Maidaneh Abdi et al, 2020). Some studies concentrate on the history of edition of an object to evaluate its quality (Barron et al, 2014).…”

“…It is about inferring an extrinsic indicator by using intrinsic indicators based on the research hypothesis that the discrepancies between VGI and reference data can be reconstructed from intrinsic indicators. Machine learning methods are used to establish the relationships between intrinsic features and the target extrinsic indicator (Mohammadi and Malek, 2015, Xu et al, 2017, Maidaneh Abdi et al, 2020. The last showed that individual intrinsic indicators gave relevant information about shape accuracy, but that they did not capture absolute accuracy very well.…”

“…For the input indicators, we use a set of fourteen intrinsic indicators, described in (Maidaneh Abdi et al, 2020). We chose to test a wide range of indicators , because we had little preconceived notion of which indicators would be the most useful, and because random forests methods are robust to the addition of indicators that carry little information.…”

“…Assessing the quality of VGI with only intrinsic characteristics, is still largely an open challenge (Barron et al, 2014, Ivanovic et al, 2019, Truong et al, 2019, and the extent to which quality measurements can be inferred from intrinsic indicators is still an open issue (Maidaneh Abdi et al, 2020, Xu et al, 2017.…”

Is the radial distance really a distance? An analysis of its properties and interest for the matching of polygon features

“…When dealing with heterogeneous GI coming from multisources such as different data models, mapping specifications, spatial and temporal scales, as well as non exhaustive or incomplete semantic and thematic information in data sources to be matched, such as for building data, deriving similarity measures that are robust to complex polygons and geometric errors becomes a challenge. Shape measures proved to be efficient for matching polygons datasets since they better distinguish the features between them (Meng and Lu 2014, Fan et al 2014, Xavier et al 2016, Maidaneh Abdi et al 2020. Thus, in the context of heterogeneous GI data matching, there is a need to describe the similarity of polygons in a very fine way.…”