USE OF VOLUNTEERED GEOGRAPHIC INFORMATION (VGI) FOR TRANSPORTATION MODELING IN RURAL AND URBAN AREAS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 0224307
• Project Start Date: Jan 1, 2011
• Project End Date: Dec 31, 2016
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF FLORIDA
G022 MCCARTY HALL
GAINESVILLE,FL 32611

Performing Department
Ft. Lauderdale Research and Education Center

Non Technical Summary
The strong demand for freely available spatial data within recent years has boosted the availability of Volunteered Geographic Information (VGI) (Goodchild 2007) on the Internet. The development of Web 2.0, the Global Positioning System (GPS) and its integration into cell phones, photo cameras, and other mobile devices, allows the Web community to interact with each other, provide information to central sites, and thus become a significant source of geographic information. VGI can be found in various Web services and other digital data sources. It includes, for example, geotagged entries in Wikipedia , place descriptions in Wikimapia , geotagged photographs in flickr , Panomario , locr , and collaborative volunteer efforts to create public domain street data layers as is done in OpenStreetMap (OSM). VGI provides a rich source of geographic information which complements geographic information that can be extracted from traditional sources, such as Earth imagery. One can see humanity as a large collection of intelligent, mobile sensors, which combine their abilities and knowledge, such as place names, with devices that collect geographic information (Goodchild 2007). Web 2.0 has arrived in our everyday life through common applications, such as Google Earth. The latter allows to combine different information sources within a Web based mapping application and visually overlay geospatial data layers using Google Earths Application Program Interface (API). Tens of thousands of sources, many of them developed by citizens with no prior experience in geographic information technologies, have taken advantage of this mechanism in recent months. These users provide overlays depicting, for example, mapped Battles and Routes of Alexander the Great , wine regions of the world, or major European ski resorts. Whereas Web 2.0 is omnipresent for many Internet users, and a tremendous amount of place-based data, images, and other geographically relevant information are now readily available, the possible applications of VGI for transportation planning have not yet received much attention in the research literature. This CRIS project focuses therefore on exploring the usability of VGI for several transportation related research topics.

Animal Health Component

(N/A)

Research Effort Categories

Basic

(N/A)

Applied

(N/A)

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
134	0530	2020	25%
134	0530	3030	25%
134	0599	2020	25%
134	0599	3030	25%

Knowledge Area
134 - Outdoor Recreation;

Subject Of Investigation
0599 - Recreational resources, general/other; 0530 - Parks and urban green space;

Field Of Science
2020 - Engineering; 3030 - Information and communication;

Keywords

vgi

geographic information

tourist routes

routing algorithm

geotagged images

transit accessibility

Goals / Objectives
The overall goal of this research is to explore whether Volunteered Geographic Information (VGI) can be used to complement Geospatial datasets available from public institutions or commercial providers in a meaningful way. This would lead to better informed spatial decision making process, and to increased knowledge about travel patterns and perception of space through VGI. Objectives: 1. Analyze the quality of Volunteered Geographic Information (VGI) for the identification of scenic locations and route segments in rural and urban areas 2. Develop a least-cost based algorithm for the computation of touristic routes 3. Use of VGI for improved modeling of pedestrian accessibility to public transit systems Expected outputs include expanded quality assessments of VGI spatial datasets, such as position accuracy of geotagged images, faster algorithms for the computation of scenic routes, and an assessment of the completeness of pedestrian related street data of proprietory and volunteered origin.

Project Methods
Procedures: Objective 1. Analyze the quality of Volunteered Geographic Information (VGI) for identification of scenic locations and route segments in rural and urban areas. This research will involve the following activities: 1.1 Extraction of footprints of geotagged images. The footprints of geotagged images for Florida will be downloaded from provided APIs on Web 2.0, including flickr, Panoramio, and locr Web sites. 1.3 Computation of fastest routes. Fastest route computation can be conducted in ArcGIS using Network Analyst for geographic regions where base street data, e.g. NAVTEQ or OSM are available. 1.4 Statistical analysis. The number of geotagged images within a given buffer along scenic and fastest route pairs are computed. A statistical test for pair wise comparison will be conducted. Procedures: Objective 2. Computation of scenic routes. This research will involve the following activities: 2.1 Analyze existing scenic routes. A first step is to identify the characteristics of scenic routes. A starting point is to analyse scenic routes posted on Web 2.0. 2.2 Formulate the problem in linear programming notation. This formulation includes the expression to be minimized or maximized (the objective function for a scenic route), and constraints 2.3 Develop a modified nearest facility search procedure.This step aims to develop a faster modified search procedure for nearest facilities based on re-using of already traversed path segments. 2.4 Find Shortest Path and exact solution, the latter using an external solver, such as CPLEX, 2.5 Statistical Analysis. This involves computation of the following: - Coefficient of elasticity for different weight factors f which describes how the scenic route portion (in %) increases in response to increased detour (in %). - Reachability of relatively isolated scenic elements - The range of detour associated with a weight multiplication factor f - Comparison of performance of SP based results to benchmark results from exact solution. - Comparison between computed and posted routes, i.e., how well Web 2.0 posted routes can be reproduced through the exact and the SP solution Procedures: Objective 3.: Use of VGI for improved modeling of pedestrian accessibility to public transit systems. This research will involve the following activities: 3.1 Compare percentage of pedestrian segments in data sets. This step analyzes for different street datasets the percentage of pedestrian-only segments for freely available street data sets, such as TIGER or OSM, and commercial data sets, such as NAVTEQ or Tele Atlas. 3.2 Compute services areas. Using a GIS and the locations of transit stops, generate service areas for different transit types and street networks. 3.3 Statistical analysis. Compute the relative growth of service areas through the integration of pedestrian-only segments. Compare the effect between different geographic areas, between different data sources, and transportation means. Analyze correlation between service area growth and density of pedestrian-only paths.

Progress 01/01/11 to 12/31/16

Outputs
Target Audience:Most of my research publications aim to reach researchers and practiotioners in transportation engineering. Besides publishing, I co-organized a workshop at an international conference about Volunteered Geographic Information in Helsinki, Finland, which was attended by 30 researchers (students and faculty). ThroughSouth Florida Career Construction Days several hundred middle-and highschool students were reached and educated about career opportunities in Geomatics. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Co-organization of an international workshop called "LINKing and analyzing Volunteered Geographic Information (VGI) across different platforms" (http://www.geog.uni-heidelberg.de/gis/link_vgi.html) provided discussion opportunities and hands-on training to 30 participants. How have the results been disseminated to communities of interest?We began to post RLOs on our Geomatics YouTube channel, which reaches a wide audience of the general public interesting in geo-mapping topics. Within the past year, we counted around 6000 views of our educational materials. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? We compared the spatio-temporal development of photo contributions to the crowd-sourced photosharing services Flickr and Panoramio using California as a case study, and could determine which data source is preferablyused in urban or rural areas. We found that Panoramio image constributions were on the decline whereas Flickr photo contributions were still on the rise. We began analyzing crowd-sourced GPS tracking data provided by Strava to analyze socioeconomic factors, road network properties, and location specific characteristics that affect bicycle ridership. This showed which features appear to be scenic to cyclists (e.g. proximity to bay or ocean). We identified limitations of data quality for this purpose. We began to analyze the cross-reference between various VGI data platforms, e.g. OSM and Mapillary, and found that users begin to combine several VGI data sources in mapping, which we expect to increase data quality in the long run.

Publications

• Type: Journal Articles Status: Published Year Published: 2016 Citation: Scheffrahn, R. H., Hochmair, H. H., Tonini, F., Krecek, J., Su, N.-Y., Fitzgerald, P., Hendricken, K., Chase, J. A., Mangold, J., and Olynik, J. (2016). Proliferation of the invasive termite Coptotermes gestroi (Isoptera: Rhinotermitidae) on Grand Cayman and overall termite diversity on the Cayman Islands. Florida Entomologist, 99(3), 496-504.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Cvetojevic, S., Juh�sz, L., and Hochmair, H. H.(2016). Positional Accuracy of Twitter and Instagram Images in Urban Environments. GI_Forum, 2016(1), 191-203.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Hochmair, H.H. (2016). Spatiotemporal Pattern Analysis of Taxi Trips in New York City.Transportation Research Record: Journal of the Transportation Research Board, 2542, 45-56.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Juh�sz, L. and Hochmair, H.H. (2016). User Contribution Patterns and Completeness Evaluation of Mapillary, a Crowdsourced Street Level Photo Service. Transactions in GIS, 20(6), 925-947.
• Type: Websites Status: Published Year Published: 2016 Citation: https://www.youtube.com/channel/UCJrKjbiWr73z3-qrxMGzsUw/videos?view=0&sort=dd&shelf_id=0
• Type: Book Chapters Status: Published Year Published: 2016 Citation: Juh�sz, L. and Hochmair, H. H. (2016). Cross-linkage between Mapillary Street Level Photos and OSM Edits. In T. Sarjakoski, M. Y. Santos and T. Sarjakoski (Eds.), Geospatial Data in a Changing World: Selected papers of the 19th AGILE Conference on Geographic Information Science (Lecture Notes in Geoinformation and Cartography) (pp. 141-156). Berlin: Springer.

Progress 10/01/14 to 09/30/15

Outputs
Target Audience:Researchers and planners using Volunteered Geographic Information as data source for various types of projects; general public interested in using and contributing crowd-sourcedspatial data Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Published research results invariousjournal papers and presented at various international GIS and transportation conferences. Ongoing research results werepresented to professional stakeholder groups, e.g. land surveyors, and the general public visitingcampus on site, e.g. during Open House events. What do you plan to do during the next reporting period to accomplish the goals?Analyze how users contribute to severalVGI data sources at the same time and how this affects VGI data quality; compare completeness/coverage of various VGI and social media data sources across the nation and assess their affect on data usability for planning of non-motorized trips

Impacts
What was accomplished under these goals? Analyzedquality for novel VGI data sources, such as Mapillary or drone imagery, and used VGI to determine travel patterns found in touristic routes. Assessed for selected cities in the US and Europe OpenStreetMapdata completeness for the purposes of cycling.

Publications

• Type: Journal Articles Status: Published Year Published: 2015 Citation: Alivand, M. and Hochmair, H. H. (2015). Choice Set Generation for Modeling Scenic Route Choice Behavior Using GIS. Transportation Research Record: Journal of the Transportation Research Board, 2495, 101-111.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Hochmair, H. H., Zielstra, D. and Neis, P. (2015). Assessing the Completeness of Bicycle Trail and Designated Lane Features in OpenStreetMap for the United States. Transactions in GIS, 19, 63-81
• Type: Journal Articles Status: Awaiting Publication Year Published: 2016 Citation: Juh�sz, L. and Hochmair, H. H. (forthcoming). User Contribution Patterns and Completeness Evaluation of Mapillary, a Crowdsourced Street Level Photo Service. Transactions in GIS.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Hochmair, H. H. and Zielstra, D. (2015). Analysing User Contribution Patterns of Drone Pictures to the dronestagram Photo Sharing Portal. Journal of Spatial Science, 60, 79-98.

Progress 10/01/13 to 09/30/14

Outputs
Target Audience: Researchers, application developers, and the general public who want to use Volunteered Geographic Information to make better informed spatial decisions, such as planning for scenic trips in urban and rural areas. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? I worked with 2 of my PhD students and one exchange M.S. student from Austria on this research, preparing on several manuscripts for peer-reviewed journals. One PhD student presented his research on Twitter data analysis at a national Geographer's conference. I employed one part-time intern to work on data digitizing tasks. How have the results been disseminated to communities of interest? I published together with my graduate students several peer-reviewed papers in journals and conference proceedings. One journal paper was published in an open access journal which allows the general public to view the publication. Further we released two newsletters informing stakeholders, potential future students, and working professionals of our research activities. What do you plan to do during the next reporting period to accomplish the goals? Continue to publish with my graduate students in peer-reviewed journals and to present research results and national and international conferences.

Impacts
What was accomplished under these goals? Several environmental attributes were found to contribute to route scenery, including water, forest, and mountains. Urban areas were identified to be negatively associated with scenery. These results can help to build better tools to plan one's vacation or leisure trip. It was also found that a combination of clustering processes on OpenStreetMap data allows to identify contributors' home regions for this important VGI data source. This was tested with 10 regularly data contributors. Such regions can be used as a proxy for increased data quality. Further, it was shown that Twitter data reveal some general travel trends of Twitter users, but it was also found that this data is too sparse to reconstruct individual travel behavior and local travel.

Publications

• Type: Journal Articles Status: Published Year Published: 2014 Citation: Zielstra, D., Hochmair, H. H., Neis, P., and Tonini, F. (2014). Areal delineation of home regions from contribution and editing patterns in OpenStreetMap. ISPRS International Journal of Geo-Information, 3(4), 1211-1233.
• Type: Book Chapters Status: Published Year Published: 2014 Citation: Hochmair, H.H., and Cvetojevic, S. (2014). Assessing the Usability of Georeferenced Tweets for the Extraction of Travel Patterns: A Case Study for Austria and Florida. In R. Vogler, A. Car, J. Strobl, and G. Griesebner (Eds.), GI_Forum 2014. Geospatial Innovation for Society (pp. 30-39). Berlin/Offenbach: Wichmann.
• Type: Journal Articles Status: Accepted Year Published: 2015 Citation: Alivand, M., Hochmair, H. H., and Srinivasan, S. (forthcoming). Analyzing how travelers choose scenic routes using route choice models. Computers, Environment and Urban Systems

Progress 10/01/12 to 09/30/13

Outputs
Target Audience: Academic and industry research community that develops geospatial products and applications based on Volunteered Geographic Information (VGI) Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Research results were published primarily in peer-reviewed journal paper and VGI related Web blogs. What do you plan to do during the next reporting period to accomplish the goals? 1) Refine regression models for determining attributes of scenic routes. 2) Assess the uability of Twitter data for the extraction of travel behavior in urban and rural areas. 3) Analyze the OSM data growth using spatio-temporal lagged regression models. 4) Analyze OSM user editiing patterns to determine a mapper's home region.

Impacts
What was accomplished under these goals? 1) We developed a method to extract scenic routes from VGI data sources using urban and rural areas of California as test area. This was presented at an international GIS workshop titled: Second ACM SIGSPATIAL International Workshop on Crowdsourced and Volunteered Geographic Information (GEOCROWD) . Further we ran regression models that identify which factors contribute to scenic routes, which showed VGI images from Panoramio as one contributing factors. This factor can and should thus be included in the computation of scenic routes. 2) I co-organized a workshop on Action and Interaction in Volunteered Geographic Information in Belgium 3) My research group published findings on VGI data quality as follows: a) We identified the spatial accuracy of VGI shared images from Panoramio and Flickr. b) We assessed the completeness of OpenStreetMap (OSM) bicycle trail and lane data in the United States. c) We analyzed how TIGER road data imports into OpenStreetMap affects the data quality of road data in OSM.

Publications

• Type: Journal Articles Status: Accepted Year Published: 2014 Citation: Hochmair, H.H., Zielstra, D., and Neis, P. (accepted). Assessing the Completeness of Bicycle Trail and Designated Lane Features in OpenStreetMap for the United States. Transactions in GIS.
• Type: Journal Articles Status: Published Year Published: 2013 Citation: Mooney, P., Rehrl, K., and Hochmair, H.H. (2013). Action and Interaction in Volunteered Geographic Information: A Workshop Review. Journal of Location Based Services, 7(4), 291-311.
• Type: Journal Articles Status: Published Year Published: 2013 Citation: Zielstra, D. and Hochmair, H.H. (2013). Positional accuracy analysis of Flickr and Panoramio images for selected world regions. Journal of Spatial Science, 58(2), 251-273.
• Type: Journal Articles Status: Published Year Published: 2013 Citation: Zielstra, D., Hochmair, H.H., and Neis, P. (2013). Assessing the Effect of Data Imports on the Completeness of OpenStreetMap - A United States Case Study. Transactions in GIS, 17(3), 315-334.

Progress 10/01/11 to 09/30/12

Outputs
OUTPUTS: I continued to analyse the data quality of Volunteered Geographic Information (VGI). This included: - comparing the completeness of free (TIGER, OSM) and proprietary data sources (NAVTEQ, TomTom) regarding pedestrian navigation - assessing completeness of of bicycle trails and designated lane features in OpenStreetMap (OSM) - assessing the positional accuracy of images in photo sharing services, e.g. Flickr and Panoramio I hosted an MSc student from Austria in spring 2013 for 3 months who continued to work on analyzing the temporal development of different feature classes in OSM. I submitted a workshop proposal Action and Interaction in Volunteered Geographic Information (ACTIVITY)" with some co-authors. )". The submitted workshop proposal was accepted by the conference program committee, and it will be held during the 16th AGILE Conference on Geographic Information Science in Leuven, Belgium (14-17 May 2013). PARTICIPANTS: Hartwig Hochmair, PI Dennis Zielstra, PhD graduate assistant TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
Panoramio images were found to me more accurate than Flickr images. Also, we found differences in spatial accuracy between different continents. This has implications when designing software that relies on adding VGI image sources, e.g. route planners with virtual tours for pre-view. The completeness of bicycle features in OSM varies strongly between different states and urban areas in the US. This means that OSM is more reliable for certain areas than for others, which affects the use of data sources in bicycle network analysis for transportation and engineering tasks.

Publications

• Rehrl, K., Groechenig, S., Hochmair, H. H., Leitinger, S., Steinmann, R., and Wagner, A. (2013). A Conceptual Model for Analyzing Contribution Patterns in the Context of VGI. In J. M. Krisp (Ed.), Progress in Location Based Services (Lecture Notes in Geoinformation and Cartography). Springer: Berlin.
• Hochmair, H. H., Zielstra, D., and Neis, P. (2013). Assessing the Completeness of Bicycle Trails and Designated Lane Features in OpenStreetMap for the United States and Europe. Proceedings of Transportation Research Board - 92nd Annual Meeting, Washington, D.C. Transportation Research Board of the National Academies.
• Zielstra, D., and Hochmair, H. H. (2013). Comparing Shortest Paths Lengths of Free and Proprietary Data for Effective Pedestrian Routing in Street Networks. Transportation Research Record: Journal of the Transportation Research Board.
• Hochmair, H.H., and Zielstra, D. (2012). Positional Accuracy of Flickr and Panoramio Images in Europe. In Jekel, T., A. Car, G. Griesebner and J. Strobl (Eds.), GI_Forum 2012: Geovisualization, Society and Learning (pp. 14-23). Berlin: Wichmann.

Progress 10/01/10 to 09/30/11

Outputs
OUTPUTS: I designed and conducted an empirical study with 20 participants which will quantify the spatial accuracy of Volunteered Geographic Information (VGI) from flickr and Panoramio point sources. Analysis of survey data is underway, with results expected to be ready by the end of 2011. This will give insight how useful VGI is for annotating route directions with VGI image data. Related to this I developed a new lecture unit on Volunteered Geographic Information in the SUR 5365 Digital Mapping course which I am currently teaching. I continue the mentoring of one PhD student (Dennis Zielstra) on the topic of quality assessment of VGI, and a second one (Majid Alivand) on the algorithmic aspects of scenic route computations. Dissemination was done through several conference presentations, such as the Annual Transportation Research Board (TRB) Meeting (Jan 2011, Washington, DC), and the International ESRI User conference (July 2011, San Diego, CA) PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
New findings show that VGI data, in particular OpenStreetMap data, are a useful and cost-free alternative to commercial datasets for pedestrian routing, especially in Europe. This finding is relevant for route planner applications that optimize route search for various optimization criteria, such as shortest or most scenic route. Representatives of public agencies attending the TRB conference presentation showed interest in utilizing the presented method for their own network analysis to assess service areas around transit stops.

Publications

• Zielstra, D. and Hochmair, H. H. (2011). A Comparative Study of Pedestrian Accessibility to Transit Stations Using Free and Proprietary Network Data. Transportation Research Record: Journal of the Transportation Research Board
• Zielstra, D., and Hochmair, H. H. (2011). Digital Street Data: Free versus Proprietary. GIM International, 25 (7), 29-33