=Paper=
{{Paper
|id=Vol-1598/paper15
|storemode=property
|title=Detection of potential updates of authoritative spatial databases by fusion of volunteered geographical information from different sources
|pdfUrl=https://ceur-ws.org/Vol-1598/paper15.pdf
|volume=Vol-1598
|authors=Stefan S. Ivanović,Ana-Maria Olteanu Raimond,Sébastien Mustière,Thomas Devogele
|dblpUrl=https://dblp.org/rec/conf/agile/IvanovicRMD15
}}
==Detection of potential updates of authoritative spatial databases by fusion of volunteered geographical information from different sources==
https://ceur-ws.org/Vol-1598/paper15.pdf
Detection of potential updates of authoritative
spatial databases by fusion of Volunteered
Geographical Information from different
sources
Stefan S. Ivanović*1, Ana-Maria Olteanu Raimond1, Sébastien Mustière1,
Thomas Devogele2
1
Université Paris-Est, IGN, COGIT Lab, France
2
Université François Rabelais de Tours, France
A continuous update of authoritative spatial databases is highly demanding task in
both aspects, technical and financial. In the same time, alternative modalities to
collect content, in particular spatial content, have achieved a certain maturity and
must be considered as they may leverage the cost of updating authoritative spatial
databases (Al-Bakri, 2010). This alternative data known as Volunteered Geo-
graphical Information – VGI (Goodchild, 2007) is easy available and is being col-
lected in almost every moment somewhere in the world.
GPS tracks, in particular, seem to be a relevant source of update information to
improve the freshness of a road network. Walkway, tractor and bicycle roads are
identified as very challenging types for continuous update due to their intermittent
nature (e.g. they appear and disappear very often) as well as various landscape
(e.g. forest, high mountains, seashore, etc.). Even though, these types of roads are
not of the highest priority for a national mapping agency (not a lot of resources are
devoted to their update), they are still very important for production of touristic
maps and for other different applications such as defense, sport activities, etc. Fur-
thermore, a connectivity of the entire network depends on them also. The main ob-
jective of this research is proposing a method for identifying potential updates of
mentioned road type of authoritative spatial databases using VGI data, more pre-
cisely GPS tracks.
Hence, we have focused on GPS traces obtained in sport activities, since they are
mainly collected along those roads. Moreover, they are widely available on the
websites of French sport association activities like RandoGPS, OpenRunner, Uta-
gawaVTT etc.
_________________________
Copyright (c) by the paper's authors. Copying permitted for private and academic purposes. In:
A. Comber, B. Bucher, S. Ivanovic (eds.): Proceedings of the 3rd AGILE Phd School, Champs
sur Marne, France, 15-17-September-2015, published at http://ceur-ws.org
2
In order to detect potential updates - the differences between authoritative and
VGI road networks, a data matching process is supposed to be applied. Matching
links will be created between same roads represented in two different datasets
(road networks).
As a result, three main situations were distinguished:
1. There is VGI trace but no corresponding IGN – a trace without
matching link
2. There is IGN trace but no corresponding VGI – a trace without
matching link
3. There are both, IGN and VGI trace – a trace without matching link
The first situation is related to the creation of a new road in the real world, which
has not been represented yet in a corresponding authoritative spatial database.
That is considered an alert for update and requires adding the road in the database.
The second situation represents a difference between authoritative and VGI data
set in the way that a road which exists in a database, does not exist anymore in the
reality. Subsequently, that road needs to be deleted from the database.
The third possible case is not a real alert for update, since there are no differences
in compared data sets. However it is useful as a confirmation of a presence of the
roads contained in the database in the reality.
However in the situations when there is more than one link either in VGI or au-
thoritative dataset, we intend to estimate the average geometry of the trace and
continue our work based on three situations described above.
Then, the question of VGI tracks quality arises. Furthermore, VGI traces are col-
lected without any specified procedures, less or inexistent metadata, usually by
low class GPS devices. Hence, heterogeneity of data is very high as well as spatial
inaccuracy. In current stage of our work we focus on examination of data quality,
especially on its spatial and temporal aspects.
First, we present an overview of VGI data sources (websites) and the heterogenei-
ties that characterize them. In terms of data, I can rely on spatiotemporal data (i.e.
coordinates and sometimes elevation and timestamps) as well as on a variety of
descriptive information in text format such as: type of activity, difficulty, trace de-
scription etc.
3
To make the most of contextual information, we perform a comprehensive analy-
sis of context elements which affects GPS data quality. Sources of errors related to
technical aspect of GPS data collection are partially important for this research.
Since we use data obtained by low class GPS receivers, which positional accuracy
is at meter level, we are not concerned about the sources that affect the accuracy at
sub-meter level. Therefore, our attention is directed to identifying and classifying
sources of errors according to which extent they affect positional accuracy of GPS
tracks.
Finally, we are interested in evaluation of data quality by analyzing VGI data itself
– Intrinsic approach (Batini & Scannapieco, 2006). Thus, I tend to obtain the more
statistical indicators of data quality that I can, such as indicators of: spatial disper-
sion, precision, reliability, correlation between data etc. As a result, a process of
automatic collection of GPS traces from web-sites and storing them into Post-
greSQL database was created, as well as a variety of tools designated to the indi-
cators calculation. Evaluation of data quality is conducted by using an open source
platform GeOxygene1, developed by COGIT laboratory. Future work will aim to
establishing a unique procedure for GPS tracks data quality evaluation.
References
Goodchild, M.F. (2007). Citizens as sensors: The world of volunteered geogra-
phy. GeoJournal, 69 (4), 211–221. doi:10.1007/s10708-007-9111-y.
Al-Bakri, M.; Fairbairn, D (2010). Assessing the Accuracy of “Crowdsourced”
Data and its Integration with Official Spatial Data Sets. In Proceedings of the
Ninth International Symposium on Spatial Accuracy Assessment in Natural Re-
sources and Environmental Sciences, Leicester, UK, 20–23 July 2010; pp. 317–
320.
Batini, C., & Scannapieco, M. (2006). Data quality, Concepts, Methodologies and
Techniques. In C. Batini & K. M. Scannapieco (Eds.), Approaches to the defini-
tion of Data Quality Dimensions: Empirical approach (pp. 39). Berlin Heidelberg,
Germany: Springer-Verlag.
1 Geoxygene : http://oxygene-project.sourceforge.net/