SlideShare a Scribd company logo

A GRASS-based automated procedure to compare OpenStreetMap and authoritative road network datasets

Marco Minghini, profile picture
Marco Minghini

The document outlines a novel automated methodology developed at Politecnico di Milano for comparing OpenStreetMap (OSM) data with authoritative road network datasets, focusing on spatial accuracy and completeness. It details the multi-step process involving preliminary comparisons, geometric preprocessing, and a grid-based evaluation of OSM accuracy, combined with sensitivity analyses. This method is designed to be flexible and extensible, facilitating comparisons across various datasets, while emphasizing the challenges of assessing volunteered geographic information (VGI) data quality.

Engineering
1 of 62
Download to read offline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
GEOlab, Como Campus A GRASS-based automated procedure to compare OpenStreetMap and authoritative road network datasets Politecnico di Milano, Como Campus, DICA, via Valleggio 11, 22100 Como (Italy) Maria Antonia Brovelli, Marco Minghini, Monia Elisa Molinari
2 ✔ Increasing popularity of OpenStreetMap (OSM) as today's most notable Volunteered Geographic Information (VGI) project on the Internet ✔ Increasing concern on VGI (and OSM) data quality: Motivation of the work – VGI & OSM quality GEOlab, Politecnico di Milano – Como Campus ➔ spatial accuracy ➔ temporal accuracy ➔ semantic accuracy ➔ up-to-dateness ➔ consistency ➔ fitness-for-use & fitness-for-purpose ➔ ... ✔ Increasing availability of open data from NMAs and CSC that can be used as a source of comparison for VGI (and OSM) data: ➔ comparing two spatial datasets against each other is a challenging geocomputation problem!
3 ✔ Literature provides plenty of works assessing or comparing OSM quality against that of authoritative datasets: Motivation of the work – OSM comparisons ➔ strongly focused on road network ➔ mainly investigating OSM positional accuracy ➔ OSM compared to data from NMA (UK Ordnance Survey, French NMA, USGS TNM/TIGER, etc.) and CSC (Navteq, TeleAtlas, etc.) ➔ semi- or fully-automated ➔ results from poor to very good ✔ Comparison techniques are very strong and fit for purpose, but mostly application and dataset specific: ➔ hard to replicate ➔ difficult to extend to other dataset comparisons GEOlab, Politecnico di Milano – Como Campus
4 ✔ Novel methodology to compare OSM and authoritative road datasets: Our methodology ➔ fully automated ➔ focused on spatial accuracy and completeness ➔ flexible, i.e. not developed for a specific dataset ➔ built with FOSS4G (Free and Open Source Software for Geospatial) ✗ made of required and optional operations ✗ users can define the value of the parameters involved to adapt the procedure to their specific authoritative datasets ✗ users are supposed to be familiar with the authoritative dataset used as reference ✗ reusable and extensible in case of need GEOlab, Politecnico di Milano – Como Campus
5 ➔ 1. Preliminary comparison of the datasets and computation of global statistics ➔ 2. Geometric preprocessing of the OSM dataset to extract a subset which is fully comparable with the IGN dataset ➔ 3. Evaluation of OSM spatial accuracy using a grid-based approach ✔ Currently developed as 3 GRASS GIS modules: Our methodology – Overview ➔ written in Python ➔ available with a Graphical User Interface (GUI) ✔ Comparison between OSM and reference (IGN) road network datasets composed of 3 consecutive steps: GEOlab, Politecnico di Milano – Como Campus
6 ✔ Import and select the OSM and IGN datasets [required] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus
7 ✔ Import and select the OSM and IGN datasets [required] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus data © IGN and © OpenStreetMap contributors
8 ✔ Import and select the OSM and IGN datasets [required] ✔ If the extent of the OSM and/or IGN datasets is larger than the one of interest, import a vector layer to be used as clipping mask [optional] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus
9 ✔ Apply a set of buffers of user-specified width around both the IGN and OSM datasets, to compute the length and the length percentage of the OSM and IGN datasets included in the buffer [required] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus
10 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus
11 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus
12 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus ✗ ≅450 km more in OSM than IGN dataset!
13 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus ✗ ≅450 km more in OSM than IGN dataset! ➔ more footways and pedestrian routes mapped in OSM Boulevard des Invalides Gare de l'Est IGN OSM data © IGN and © OpenStreetMap contributors
14 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus ✗ ≅450 km more in OSM than IGN dataset! ➔ cycleways and carriageways mapped as separate highways in OSM IGN OSM Boulevard Jules Ferry data © IGN and © OpenStreetMap contributors Boulevard Henri IV
15 ✔ Outputs from Step 1 can be used to perform further analysis: Step 1: Preliminary comparison of the datasets ➔ sensitivity analysis on the buffer width GEOlab, Politecnico di Milano – Como Campus
16 ✔ Outputs from Step 1 can be used to perform further analysis: Step 1: Preliminary comparison of the datasets ➔ sensitivity analysis on the buffer width GEOlab, Politecnico di Milano – Como Campus
17 ✔ Cleaning of OSM dataset to make it comparable with IGN dataset Step 2: preprocessing of the OSM dataset GEOlab, Politecnico di Milano – Como Campus
18 ✔ Cleaning of OSM dataset to make it comparable with IGN dataset Step 2: preprocessing of the OSM dataset ➔ computationally intensive – work area divided in 4 sub-areas GEOlab, Politecnico di Milano – Como Campus 1 4 2 3 data © IGN and © OpenStreetMap contributors
19 ✔ Generalize the IGN dataset with the Douglas-Peucker algorithm [optional] Step 2: preprocessing of the OSM dataset ➔ users have to enter the threshold for the algorithm GEOlab, Politecnico di Milano – Como Campus
20 ✔ Generalize the IGN dataset with the Douglas-Peucker algorithm [optional] Step 2: preprocessing of the OSM dataset ✔ Split the line features of the datasets into segments [required] ➔ users have to enter the threshold for the algorithm GEOlab, Politecnico di Milano – Como Campus
21 ✔ Compute a measure of degree for the nodes of IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ identify the terminal nodes (degree = 1) GEOlab, Politecnico di Milano – Como Campus data © IGN
22 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer GEOlab, Politecnico di Milano – Como Campus
23 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer ➔ buffer is applied without cap around the terminal nodes GEOlab, Politecnico di Milano – Como Campus
24 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer ➔ buffer is applied without cap around the terminal nodes GEOlab, Politecnico di Milano – Como Campus
25 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer ➔ buffer is applied without cap around the terminal nodes GEOlab, Politecnico di Milano – Como Campus
26 Step 2: preprocessing of the OSM dataset ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
27 Step 2: preprocessing of the OSM dataset ➔ apply a buffer of user-specified width around each IGN segment ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
28 Step 2: preprocessing of the OSM dataset ➔ compute the angular coefficient of each IGN segment and all the OSM segments included in the buffer around it ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
29 Step 2: preprocessing of the OSM dataset ➔ compare the difference between IGN and OSM angular coefficients with a user-specified threshold ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
30 Step 2: preprocessing of the OSM dataset ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
31 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset ➔ sensitivity analysis on the parameters involved GEOlab, Politecnico di Milano – Como Campus data © IGN and © OpenStreetMap contributors
32 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset ➔ sensitivity analysis on the parameters involved GEOlab, Politecnico di Milano – Como Campus
33 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset GEOlab, Politecnico di Milano – Como Campus ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m
34 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m GEOlab, Politecnico di Milano – Como Campus ✗ preprocessed OSM has 50 km less than original OSM≅ ✗ preprocessed OSM has still 50 km more than IGN≅
35 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset GEOlab, Politecnico di Milano – Como Campus
36 ✔ Use a grid to take into account OSM heterogeneous nature [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ import a vector layer to be used as grid GEOlab, Politecnico di Milano – Como Campus
37 ✔ Use a grid to take into account OSM heterogeneous nature [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ import a vector layer to be used as grid ➔ manually create a grid GEOlab, Politecnico di Milano – Como Campus
38 ✔ For each grid cell, find the OSM maximum deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ enter an upper bound value for the deviation, and the percentage of OSM road length to be considered (to take into account outliers) GEOlab, Politecnico di Milano – Como Campus
39 ✔ For each grid cell, find the OSM maximum deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m 5 - 6 m 6 - 7 m 7 - 8 m 8 - 9 m 9 - 10 m 10 - 11 m
40 ✔ For each grid cell, find the OSM maximum deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m ➔ worst results are mainly due to: ✗ presence of 2 or more OSM roads for a single IGN road ✗ inherent complexity of the road network IGN OSM data © IGN and © OpenStreetMap contributors
41 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ users have to enter one or more thresholds for deviation GEOlab, Politecnico di Milano – Como Campus
42 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ length percentage of OSM roads included in the threshold buffer ➔ Area 2: threshold buffer = 6 m GEOlab, Politecnico di Milano – Como Campus 85 - 90% 90 - 95% 95 - 100%
43 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ length percentage of OSM roads included in the threshold buffer ➔ Area 2: threshold buffer = 8 m GEOlab, Politecnico di Milano – Como Campus 85 - 90% 90 - 95% 95 - 100%
44 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ length percentage of OSM roads included in the threshold buffer ➔ Area 2: threshold buffer = 10 m GEOlab, Politecnico di Milano – Como Campus 85 - 90% 90 - 95% 95 - 100%
45 ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus 0 m 0.5 m 1 m 6 m Area 1 10 10 10 Area 2 14 14 14 Area 3 7 7 7 Area 4 8 8 8 8 m Area 1 22 22 22 Area 2 26 26 26 Area 3 37 37 37 Area 4 26 26 26 10 m Area 1 28 28 28 Area 2 28 28 28 Area 3 44 44 44 Area 4 30 30 30 ➔ number of grid cells where the percentage of OSM length satisfying the given accuracy is > 95%
46 ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus 0 m 0.5 m 1 m 6 m Area 1 10 10 10 Area 2 14 14 14 Area 3 7 7 7 Area 4 8 8 8 8 m Area 1 22 22 22 Area 2 26 26 26 Area 3 37 37 37 Area 4 26 26 26 10 m Area 1 28 28 28 Area 2 28 28 28 Area 3 44 44 44 Area 4 30 30 30 ➔ number of grid cells where the percentage of OSM length satisfying the given accuracy is > 95% ➔ no difference at all!
47 Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus media dev. st. min(abs) max(abs) 6 m -0.010 0.114 0.001 0.510 8 m 0.003 0.080 0.001 0.276 10 m 0.013 0.059 0.001 0.208 ➔ statistics on the differences between the percentages of OSM length satisfying each given accuracy, for the generalization thresholds of 0 m and 1 m (Area 1) ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold:
48 ➔ generalization (within the nominal accuracy of the dataset) does not influence accuracy evaluation results – and allows to save much time! Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus ➔ statistics on the differences between the percentages of OSM length satisfying each given accuracy, for the generalization thresholds of 0 m and 1 m (Area 1) ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold: media dev. st. min(abs) max(abs) 6 m -0.010 0.114 0.001 0.510 8 m 0.003 0.080 0.001 0.276 10 m 0.013 0.059 0.001 0.208
49 ✔ Work in progress, currently available just for Step 1 Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus ➔ available at http://131.175.143.84/WPS
50 ✔ User instructions on how to use the tool Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
51 ✔ Geocoding service to move the map to a specified location Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
52 ✔ Upload of IGN road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
53 ✔ Visualization of IGN road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
54 ✔ Upload of OSM road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
55 ✔ Visualization of OSM road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
56 ✔ Definition of layers and buffer value for comparison, a PDF is generated Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
57 ✔ Retrieval of OSM road network dataset from the current map view Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
58 ✔ Retrieval of OSM road network dataset from the current map view Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
59 Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus ✔ Retrieval of OSM road network dataset from a rectangle drawn on the map or from the bounding box of an uploaded layer
60 ✔ Methodology for comparing OSM and authoritative road datasets: Conclusions ✔ Future work: ➔ understand the influence of parameters through sensitivity analysis ➔ reduce computational time (especially for Step 2) ➔ increase usability through a WPS implementation (also for Steps 2 & 3) ➔ extend the procedure to also compare attributes ➔ test on different datasets (any test dataset is welcome)! ➔ (hopefully) fills a gap in literature ➔ not tied to a specific reference dataset ➔ generic, flexible and adaptable to any reference dataset ➔ users have a key role in driving the procedure ➔ parameter values should reflect the characteristic of the reference dataset involved (e.g. nominal scale and accuracy) GEOlab, Politecnico di Milano – Como Campus
61 Links & publications GEOlab, Politecnico di Milano – Como Campus ✔ Related publications: ➔ Brovelli M. A., Minghini M., Molinari M. and Mooney P (in press). Towards an automated comparison of OpenStreetMap with authoritative road datasets. Transactions in GIS. ➔ Antunes F., Fonte C. C., Brovelli M. A., Minghini M., Molinari M. and Mooney P. (2015) Assessing OSM Road Positional Quality with Authoritative Data. Proceedings of the VIII Conferência Nacional de Cartografia e Geodesia, Lisbon (Portugal), October 29-30, 2015. ➔ Brovelli M. A., Minghini M., Molinari M. and Mooney P. (2015) A FOSS4G-based procedure to compare OpenStreetMap and authoritative road network datasets. Geomatics Workbooks 12, 235-238, ISSN 1591-092X. ✔ Links: ➔ source code:https://github.com/MoniaMolinari/OSM-roads-comparison ➔ WPS client: http://131.175.143.84/WPS
62 Contacts Thanks for your attention! Politecnico di Milano GEOlab – Polo Territoriale di Como Via Valleggio 11, 22100 Como (Italy) marco.minghini@polimi.it @MarcoMinghini Marco Minghini GEOlab, Politecnico di Milano – Como Campus

More Related Content

PDF
A GRASS-based procedure to compare OSM and IGN Paris road network datasets
Marco Minghini
 
PDF
Qualità dei dati OpenStreetMap: sperimentazioni sulla città di Milano e risul...
Marco Minghini
 
PDF
FOSS4G in Europe; Italy and the Politecnico de Milano
Carolina Arias Muñoz
 
PDF
A new module in ADMS-Roads for elevated roads
IES / IAQM
 
PDF
Roadside barriers - Accounting for the effect of vehicle induced momentum and...
IES / IAQM
 
PDF
Uses of vehicle emissions remote sensing data for emission factor development
IES / IAQM
 
PDF
Modelling the Notre-Dame de Paris fire
ARIANET
 
PDF
Light vehicle dynamics and NOx emissions on the motorway network
IES / IAQM
 
A GRASS-based procedure to compare OSM and IGN Paris road network datasets
Marco Minghini
 
Qualità dei dati OpenStreetMap: sperimentazioni sulla città di Milano e risul...
Marco Minghini
 
FOSS4G in Europe; Italy and the Politecnico de Milano
Carolina Arias Muñoz
 
A new module in ADMS-Roads for elevated roads
IES / IAQM
 
Roadside barriers - Accounting for the effect of vehicle induced momentum and...
IES / IAQM
 
Uses of vehicle emissions remote sensing data for emission factor development
IES / IAQM
 
Modelling the Notre-Dame de Paris fire
ARIANET
 
Light vehicle dynamics and NOx emissions on the motorway network
IES / IAQM
 

Similar to A GRASS-based automated procedure to compare OpenStreetMap and authoritative road network datasets (20)

PDF
Lesson2 esa summer_school_brovelli
Maria Antonia Brovelli
 
PDF
HIC2014
Maria Antonia Brovelli
 
PDF
HIC_2014
Maria Antonia Brovelli
 
PDF
3D GeoInfo Conference
Maria Antonia Brovelli
 
PDF
SFScon16 - Gianluca Antonacci: "The CLEAN-ROADS project case study"
South Tyrol Free Software Conference
 
PDF
Lesson3 esa summer_school_brovelli
Maria Antonia Brovelli
 
PDF
NextGEOSS: The Next Generation European Data Hub and Cloud Platform for Earth...
Wolfgang Ksoll
 
PDF
Systematic Approch to a GIS
Elizabeth Runkle
 
PPTX
Placing Images with Refined Language Models and Similarity Search with PCA-re...
Symeon Papadopoulos
 
PDF
Analysis of Educational Robotics activities using a machine learning approach
Lorenzo Cesaretti
 
PDF
How to Visualize the ​Spatio-Temporal Data Using CesiumJS​
SANGHEE SHIN
 
PDF
Accès ouvert aux données météorologiques d’Environnement Canada
VisionGEOMATIQUE2014
 
PPT
Francisco J. Doblas-Big Data y cambio climático
Fundación Ramón Areces
 
PDF
MediaEval 2016 - Placing Images with Refined Language Models and Similarity S...
multimediaeval
 
PDF
Frank van diggelen keynote, android gnss measurements update
The European GNSS Agency (GSA)
 
PDF
An IMU-based Turn Prediction System NOMS 2018
Yu-Chang Ho
 
PDF
Learning global pooling operators in deep neural networks for image retrieval...
Erlangen Artificial Intelligence & Machine Learning Meetup
 
PDF
Mapathon polimi 2015-11-19
Marco Minghini
 
PDF
EPANET in gvSIG
silli
 
PDF
Real time traffic management - challenges and solutions
Institute for Transport Studies (ITS)
 
Lesson2 esa summer_school_brovelli
Maria Antonia Brovelli
 
HIC2014
Maria Antonia Brovelli
 
HIC_2014
Maria Antonia Brovelli
 
3D GeoInfo Conference
Maria Antonia Brovelli
 
SFScon16 - Gianluca Antonacci: "The CLEAN-ROADS project case study"
South Tyrol Free Software Conference
 
Lesson3 esa summer_school_brovelli
Maria Antonia Brovelli
 
NextGEOSS: The Next Generation European Data Hub and Cloud Platform for Earth...
Wolfgang Ksoll
 
Systematic Approch to a GIS
Elizabeth Runkle
 
Placing Images with Refined Language Models and Similarity Search with PCA-re...
Symeon Papadopoulos
 
Analysis of Educational Robotics activities using a machine learning approach
Lorenzo Cesaretti
 
How to Visualize the ​Spatio-Temporal Data Using CesiumJS​
SANGHEE SHIN
 
Accès ouvert aux données météorologiques d’Environnement Canada
VisionGEOMATIQUE2014
 
Francisco J. Doblas-Big Data y cambio climático
Fundación Ramón Areces
 
MediaEval 2016 - Placing Images with Refined Language Models and Similarity S...
multimediaeval
 
Frank van diggelen keynote, android gnss measurements update
The European GNSS Agency (GSA)
 
An IMU-based Turn Prediction System NOMS 2018
Yu-Chang Ho
 
Learning global pooling operators in deep neural networks for image retrieval...
Erlangen Artificial Intelligence & Machine Learning Meetup
 
Mapathon polimi 2015-11-19
Marco Minghini
 
EPANET in gvSIG
silli
 
Real time traffic management - challenges and solutions
Institute for Transport Studies (ITS)
 
Ad

More from Marco Minghini (7)

PDF
Preliminary Assessment of the Global Urban Footprint and the Global Human Set...
Marco Minghini
 
PDF
OpenStreetMap a servizio delle emergenze
Marco Minghini
 
PDF
MIGRATE – MIGRation pATterns in Europe
Marco Minghini
 
PDF
MIGRATE: MIGRation pATterns in Europe
Marco Minghini
 
PDF
Building Land Use/Land Cover maps from OpenStreetMap
Marco Minghini
 
PDF
How to organize and run your own OSM humanitarian mapathon
Marco Minghini
 
PDF
The paths of Regina: Web Mapping to promote slow tourism
Marco Minghini
 
Preliminary Assessment of the Global Urban Footprint and the Global Human Set...
Marco Minghini
 
OpenStreetMap a servizio delle emergenze
Marco Minghini
 
MIGRATE – MIGRation pATterns in Europe
Marco Minghini
 
MIGRATE: MIGRation pATterns in Europe
Marco Minghini
 
Building Land Use/Land Cover maps from OpenStreetMap
Marco Minghini
 
How to organize and run your own OSM humanitarian mapathon
Marco Minghini
 
The paths of Regina: Web Mapping to promote slow tourism
Marco Minghini
 
Ad

Recently uploaded (20)

PDF
Evaluating the Democratization of the Turkish Armed Forces from a Normative P...
jpsjournal1
 
PPTX
Construction Project Organization Group 2.pptx
vj5agdales
 
PDF
BMEC211 - INTRODUCTION TO MECHATRONICS-1.pdf
ryankakungu
 
PPTX
Internet of Things (IOT) - A guide to understanding
Davies Chacko
 
PDF
Enhancing Cyber Defense Against Zero-Day Attacks using Ensemble Neural Networks
IJCNCJournal
 
PPTX
UNIT-1 - COAL BASED THERMAL POWER PLANTS
Chandra Kumar S
 
PPTX
Engineering Ethics, Safety and Environment [Autosaved] (1).pptx
MehrabTaseenTalukder
 
PDF
composite construction of structures.pdf
AinieButt1
 
PPTX
MET 305 2019 SCHEME MODULE 2 COMPLETE.pptx
VinayB68
 
PPTX
additive manufacturing of ss316l using mig welding
premcdr7799
 
PDF
Mitigating Risks through Effective Management for Enhancing Organizational Pe...
IJAEMSJORNAL
 
PPTX
CARTOGRAPHY AND GEOINFORMATION VISUALIZATION chapter1 NPTE (2).pptx
abhi1361yadav
 
PPTX
Lecture Notes Electrical Wiring System Components
eedgecbhojpur
 
PPTX
FINAL REVIEW FOR COPD DIANOSIS FOR PULMONARY DISEASE.pptx
rubeshbme
 
PDF
Operating System & Kernel Study Guide-1 - converted.pdf
mranoninvisib16
 
PDF
PPT on Performance Review to get promotions
prashant593122
 
DOCX
573137875-Attendance-Management-System-original
AYUSHMANAV
 
PPTX
Artificial Intelligence
dikshendrasarpate2
 
PPTX
OOP with Java - Java Introduction (Basics)
Rashmi T V
 
PDF
keyrequirementskkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk
mojeeburahmanwardak
 
Evaluating the Democratization of the Turkish Armed Forces from a Normative P...
jpsjournal1
 
Construction Project Organization Group 2.pptx
vj5agdales
 
BMEC211 - INTRODUCTION TO MECHATRONICS-1.pdf
ryankakungu
 
Internet of Things (IOT) - A guide to understanding
Davies Chacko
 
Enhancing Cyber Defense Against Zero-Day Attacks using Ensemble Neural Networks
IJCNCJournal
 
UNIT-1 - COAL BASED THERMAL POWER PLANTS
Chandra Kumar S
 
Engineering Ethics, Safety and Environment [Autosaved] (1).pptx
MehrabTaseenTalukder
 
composite construction of structures.pdf
AinieButt1
 
MET 305 2019 SCHEME MODULE 2 COMPLETE.pptx
VinayB68
 
additive manufacturing of ss316l using mig welding
premcdr7799
 
Mitigating Risks through Effective Management for Enhancing Organizational Pe...
IJAEMSJORNAL
 
CARTOGRAPHY AND GEOINFORMATION VISUALIZATION chapter1 NPTE (2).pptx
abhi1361yadav
 
Lecture Notes Electrical Wiring System Components
eedgecbhojpur
 
FINAL REVIEW FOR COPD DIANOSIS FOR PULMONARY DISEASE.pptx
rubeshbme
 
Operating System & Kernel Study Guide-1 - converted.pdf
mranoninvisib16
 
PPT on Performance Review to get promotions
prashant593122
 
573137875-Attendance-Management-System-original
AYUSHMANAV
 
Artificial Intelligence
dikshendrasarpate2
 
OOP with Java - Java Introduction (Basics)
Rashmi T V
 
keyrequirementskkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk
mojeeburahmanwardak
 

A GRASS-based automated procedure to compare OpenStreetMap and authoritative road network datasets

  • 1. GEOlab, Como Campus A GRASS-based automated procedure to compare OpenStreetMap and authoritative road network datasets Politecnico di Milano, Como Campus, DICA, via Valleggio 11, 22100 Como (Italy) Maria Antonia Brovelli, Marco Minghini, Monia Elisa Molinari
  • 2. 2 ✔ Increasing popularity of OpenStreetMap (OSM) as today's most notable Volunteered Geographic Information (VGI) project on the Internet ✔ Increasing concern on VGI (and OSM) data quality: Motivation of the work – VGI & OSM quality GEOlab, Politecnico di Milano – Como Campus ➔ spatial accuracy ➔ temporal accuracy ➔ semantic accuracy ➔ up-to-dateness ➔ consistency ➔ fitness-for-use & fitness-for-purpose ➔ ... ✔ Increasing availability of open data from NMAs and CSC that can be used as a source of comparison for VGI (and OSM) data: ➔ comparing two spatial datasets against each other is a challenging geocomputation problem!
  • 3. 3 ✔ Literature provides plenty of works assessing or comparing OSM quality against that of authoritative datasets: Motivation of the work – OSM comparisons ➔ strongly focused on road network ➔ mainly investigating OSM positional accuracy ➔ OSM compared to data from NMA (UK Ordnance Survey, French NMA, USGS TNM/TIGER, etc.) and CSC (Navteq, TeleAtlas, etc.) ➔ semi- or fully-automated ➔ results from poor to very good ✔ Comparison techniques are very strong and fit for purpose, but mostly application and dataset specific: ➔ hard to replicate ➔ difficult to extend to other dataset comparisons GEOlab, Politecnico di Milano – Como Campus
  • 4. 4 ✔ Novel methodology to compare OSM and authoritative road datasets: Our methodology ➔ fully automated ➔ focused on spatial accuracy and completeness ➔ flexible, i.e. not developed for a specific dataset ➔ built with FOSS4G (Free and Open Source Software for Geospatial) ✗ made of required and optional operations ✗ users can define the value of the parameters involved to adapt the procedure to their specific authoritative datasets ✗ users are supposed to be familiar with the authoritative dataset used as reference ✗ reusable and extensible in case of need GEOlab, Politecnico di Milano – Como Campus
  • 5. 5 ➔ 1. Preliminary comparison of the datasets and computation of global statistics ➔ 2. Geometric preprocessing of the OSM dataset to extract a subset which is fully comparable with the IGN dataset ➔ 3. Evaluation of OSM spatial accuracy using a grid-based approach ✔ Currently developed as 3 GRASS GIS modules: Our methodology – Overview ➔ written in Python ➔ available with a Graphical User Interface (GUI) ✔ Comparison between OSM and reference (IGN) road network datasets composed of 3 consecutive steps: GEOlab, Politecnico di Milano – Como Campus
  • 6. 6 ✔ Import and select the OSM and IGN datasets [required] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus
  • 7. 7 ✔ Import and select the OSM and IGN datasets [required] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus data © IGN and © OpenStreetMap contributors
  • 8. 8 ✔ Import and select the OSM and IGN datasets [required] ✔ If the extent of the OSM and/or IGN datasets is larger than the one of interest, import a vector layer to be used as clipping mask [optional] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus
  • 9. 9 ✔ Apply a set of buffers of user-specified width around both the IGN and OSM datasets, to compute the length and the length percentage of the OSM and IGN datasets included in the buffer [required] Step 1: Preliminary comparison of the datasets GEOlab, Politecnico di Milano – Como Campus
  • 10. 10 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus
  • 11. 11 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus
  • 12. 12 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus ✗ ≅450 km more in OSM than IGN dataset!
  • 13. 13 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus ✗ ≅450 km more in OSM than IGN dataset! ➔ more footways and pedestrian routes mapped in OSM Boulevard des Invalides Gare de l'Est IGN OSM data © IGN and © OpenStreetMap contributors
  • 14. 14 ✔ Compute also the total length of OSM and IGN datasets and their length difference, both in map units and percentage [required] Step 1: Preliminary comparison of the datasets ➔ output values are returned in a text file GEOlab, Politecnico di Milano – Como Campus ✗ ≅450 km more in OSM than IGN dataset! ➔ cycleways and carriageways mapped as separate highways in OSM IGN OSM Boulevard Jules Ferry data © IGN and © OpenStreetMap contributors Boulevard Henri IV
  • 15. 15 ✔ Outputs from Step 1 can be used to perform further analysis: Step 1: Preliminary comparison of the datasets ➔ sensitivity analysis on the buffer width GEOlab, Politecnico di Milano – Como Campus
  • 16. 16 ✔ Outputs from Step 1 can be used to perform further analysis: Step 1: Preliminary comparison of the datasets ➔ sensitivity analysis on the buffer width GEOlab, Politecnico di Milano – Como Campus
  • 17. 17 ✔ Cleaning of OSM dataset to make it comparable with IGN dataset Step 2: preprocessing of the OSM dataset GEOlab, Politecnico di Milano – Como Campus
  • 18. 18 ✔ Cleaning of OSM dataset to make it comparable with IGN dataset Step 2: preprocessing of the OSM dataset ➔ computationally intensive – work area divided in 4 sub-areas GEOlab, Politecnico di Milano – Como Campus 1 4 2 3 data © IGN and © OpenStreetMap contributors
  • 19. 19 ✔ Generalize the IGN dataset with the Douglas-Peucker algorithm [optional] Step 2: preprocessing of the OSM dataset ➔ users have to enter the threshold for the algorithm GEOlab, Politecnico di Milano – Como Campus
  • 20. 20 ✔ Generalize the IGN dataset with the Douglas-Peucker algorithm [optional] Step 2: preprocessing of the OSM dataset ✔ Split the line features of the datasets into segments [required] ➔ users have to enter the threshold for the algorithm GEOlab, Politecnico di Milano – Como Campus
  • 21. 21 ✔ Compute a measure of degree for the nodes of IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ identify the terminal nodes (degree = 1) GEOlab, Politecnico di Milano – Como Campus data © IGN
  • 22. 22 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer GEOlab, Politecnico di Milano – Como Campus
  • 23. 23 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer ➔ buffer is applied without cap around the terminal nodes GEOlab, Politecnico di Milano – Como Campus
  • 24. 24 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer ➔ buffer is applied without cap around the terminal nodes GEOlab, Politecnico di Milano – Como Campus
  • 25. 25 ✔ Apply a buffer of user-specified width around the IGN dataset [required] Step 2: preprocessing of the OSM dataset ➔ suitable buffer width derived from Step 1 ➔ delete all the OSM roads falling outside the buffer ➔ buffer is applied without cap around the terminal nodes GEOlab, Politecnico di Milano – Como Campus
  • 26. 26 Step 2: preprocessing of the OSM dataset ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
  • 27. 27 Step 2: preprocessing of the OSM dataset ➔ apply a buffer of user-specified width around each IGN segment ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
  • 28. 28 Step 2: preprocessing of the OSM dataset ➔ compute the angular coefficient of each IGN segment and all the OSM segments included in the buffer around it ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
  • 29. 29 Step 2: preprocessing of the OSM dataset ➔ compare the difference between IGN and OSM angular coefficients with a user-specified threshold ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
  • 30. 30 Step 2: preprocessing of the OSM dataset ✔ Further clean the OSM dataset [required]: GEOlab, Politecnico di Milano – Como Campus
  • 31. 31 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset ➔ sensitivity analysis on the parameters involved GEOlab, Politecnico di Milano – Como Campus data © IGN and © OpenStreetMap contributors
  • 32. 32 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset ➔ sensitivity analysis on the parameters involved GEOlab, Politecnico di Milano – Como Campus
  • 33. 33 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset GEOlab, Politecnico di Milano – Como Campus ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m
  • 34. 34 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m GEOlab, Politecnico di Milano – Como Campus ✗ preprocessed OSM has 50 km less than original OSM≅ ✗ preprocessed OSM has still 50 km more than IGN≅
  • 35. 35 ✔ Outputs from Step 2 are saved and can be used for further analysis: Step 2: preprocessing of the OSM dataset GEOlab, Politecnico di Milano – Como Campus
  • 36. 36 ✔ Use a grid to take into account OSM heterogeneous nature [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ import a vector layer to be used as grid GEOlab, Politecnico di Milano – Como Campus
  • 37. 37 ✔ Use a grid to take into account OSM heterogeneous nature [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ import a vector layer to be used as grid ➔ manually create a grid GEOlab, Politecnico di Milano – Como Campus
  • 38. 38 ✔ For each grid cell, find the OSM maximum deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ enter an upper bound value for the deviation, and the percentage of OSM road length to be considered (to take into account outliers) GEOlab, Politecnico di Milano – Como Campus
  • 39. 39 ✔ For each grid cell, find the OSM maximum deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m 5 - 6 m 6 - 7 m 7 - 8 m 8 - 9 m 9 - 10 m 10 - 11 m
  • 40. 40 ✔ For each grid cell, find the OSM maximum deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus ➔ Area 2: generalization threshold = 0.5 m, buffer = 11 m ➔ worst results are mainly due to: ✗ presence of 2 or more OSM roads for a single IGN road ✗ inherent complexity of the road network IGN OSM data © IGN and © OpenStreetMap contributors
  • 41. 41 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ users have to enter one or more thresholds for deviation GEOlab, Politecnico di Milano – Como Campus
  • 42. 42 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ length percentage of OSM roads included in the threshold buffer ➔ Area 2: threshold buffer = 6 m GEOlab, Politecnico di Milano – Como Campus 85 - 90% 90 - 95% 95 - 100%
  • 43. 43 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ length percentage of OSM roads included in the threshold buffer ➔ Area 2: threshold buffer = 8 m GEOlab, Politecnico di Milano – Como Campus 85 - 90% 90 - 95% 95 - 100%
  • 44. 44 ✔ For each grid cell, evaluate OSM accuracy against one or more threshold values of OSM deviation from IGN [optional]: Step 3: grid-based evaluation of OSM accuracy ➔ length percentage of OSM roads included in the threshold buffer ➔ Area 2: threshold buffer = 10 m GEOlab, Politecnico di Milano – Como Campus 85 - 90% 90 - 95% 95 - 100%
  • 45. 45 ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus 0 m 0.5 m 1 m 6 m Area 1 10 10 10 Area 2 14 14 14 Area 3 7 7 7 Area 4 8 8 8 8 m Area 1 22 22 22 Area 2 26 26 26 Area 3 37 37 37 Area 4 26 26 26 10 m Area 1 28 28 28 Area 2 28 28 28 Area 3 44 44 44 Area 4 30 30 30 ➔ number of grid cells where the percentage of OSM length satisfying the given accuracy is > 95%
  • 46. 46 ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold: Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus 0 m 0.5 m 1 m 6 m Area 1 10 10 10 Area 2 14 14 14 Area 3 7 7 7 Area 4 8 8 8 8 m Area 1 22 22 22 Area 2 26 26 26 Area 3 37 37 37 Area 4 26 26 26 10 m Area 1 28 28 28 Area 2 28 28 28 Area 3 44 44 44 Area 4 30 30 30 ➔ number of grid cells where the percentage of OSM length satisfying the given accuracy is > 95% ➔ no difference at all!
  • 47. 47 Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus media dev. st. min(abs) max(abs) 6 m -0.010 0.114 0.001 0.510 8 m 0.003 0.080 0.001 0.276 10 m 0.013 0.059 0.001 0.208 ➔ statistics on the differences between the percentages of OSM length satisfying each given accuracy, for the generalization thresholds of 0 m and 1 m (Area 1) ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold:
  • 48. 48 ➔ generalization (within the nominal accuracy of the dataset) does not influence accuracy evaluation results – and allows to save much time! Step 3: grid-based evaluation of OSM accuracy GEOlab, Politecnico di Milano – Como Campus ➔ statistics on the differences between the percentages of OSM length satisfying each given accuracy, for the generalization thresholds of 0 m and 1 m (Area 1) ✔ Sensitivity analysis on the Douglas-Peucker generalization threshold: media dev. st. min(abs) max(abs) 6 m -0.010 0.114 0.001 0.510 8 m 0.003 0.080 0.001 0.276 10 m 0.013 0.059 0.001 0.208
  • 49. 49 ✔ Work in progress, currently available just for Step 1 Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus ➔ available at http://131.175.143.84/WPS
  • 50. 50 ✔ User instructions on how to use the tool Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
  • 51. 51 ✔ Geocoding service to move the map to a specified location Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
  • 52. 52 ✔ Upload of IGN road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
  • 53. 53 ✔ Visualization of IGN road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
  • 54. 54 ✔ Upload of OSM road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
  • 55. 55 ✔ Visualization of OSM road network dataset Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
  • 56. 56 ✔ Definition of layers and buffer value for comparison, a PDF is generated Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus data © IGN
  • 57. 57 ✔ Retrieval of OSM road network dataset from the current map view Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
  • 58. 58 ✔ Retrieval of OSM road network dataset from the current map view Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus
  • 59. 59 Transposition of the procedure as a WPS GEOlab, Politecnico di Milano – Como Campus ✔ Retrieval of OSM road network dataset from a rectangle drawn on the map or from the bounding box of an uploaded layer
  • 60. 60 ✔ Methodology for comparing OSM and authoritative road datasets: Conclusions ✔ Future work: ➔ understand the influence of parameters through sensitivity analysis ➔ reduce computational time (especially for Step 2) ➔ increase usability through a WPS implementation (also for Steps 2 & 3) ➔ extend the procedure to also compare attributes ➔ test on different datasets (any test dataset is welcome)! ➔ (hopefully) fills a gap in literature ➔ not tied to a specific reference dataset ➔ generic, flexible and adaptable to any reference dataset ➔ users have a key role in driving the procedure ➔ parameter values should reflect the characteristic of the reference dataset involved (e.g. nominal scale and accuracy) GEOlab, Politecnico di Milano – Como Campus
  • 61. 61 Links & publications GEOlab, Politecnico di Milano – Como Campus ✔ Related publications: ➔ Brovelli M. A., Minghini M., Molinari M. and Mooney P (in press). Towards an automated comparison of OpenStreetMap with authoritative road datasets. Transactions in GIS. ➔ Antunes F., Fonte C. C., Brovelli M. A., Minghini M., Molinari M. and Mooney P. (2015) Assessing OSM Road Positional Quality with Authoritative Data. Proceedings of the VIII Conferência Nacional de Cartografia e Geodesia, Lisbon (Portugal), October 29-30, 2015. ➔ Brovelli M. A., Minghini M., Molinari M. and Mooney P. (2015) A FOSS4G-based procedure to compare OpenStreetMap and authoritative road network datasets. Geomatics Workbooks 12, 235-238, ISSN 1591-092X. ✔ Links: ➔ source code:https://github.com/MoniaMolinari/OSM-roads-comparison ➔ WPS client: http://131.175.143.84/WPS
  • 62. 62 Contacts Thanks for your attention! Politecnico di Milano GEOlab – Polo Territoriale di Como Via Valleggio 11, 22100 Como (Italy) marco.minghini@polimi.it @MarcoMinghini Marco Minghini GEOlab, Politecnico di Milano – Como Campus