Adapted from Wikipedia: OpenStreetMap (OSM) is a collaborative project to create a free editable map of the world. Created in 2004, it was inspired by the success of Wikipedia and more than two million registered users who can add data by manual survey, GPS devices, aerial photography, and other free sources. We've made available a number of tables (explained in detail below): history_* tables: full history of OSM objects planet_* tables: snapshot of current OSM objects as of Nov 2019 The history_* and planet_* table groups are composed of node, way, relation, and changeset tables. These contain the primary OSM data types and an additional changeset corresponding to OSM edits for convenient access. These objects are encoded using the BigQuery GEOGRAPHY data type so that they can be operated upon with the built-in geography functions to perform geometry and feature selection, additional processing. Example analyses are given below. This dataset is part of a larger effort to make data available in BigQuery through the Google Cloud Public Datasets program . OSM itself is produced as a public good by volunteers, and there are no guarantees about data quality. Interested in learning more about how these data were brought into BigQuery and how you can use them? Check out the sample queries below to get started. This public dataset is hosted in Google BigQuery and is included in BigQuery's 1TB/mo of free tier processing. This means that each user receives 1TB of free BigQuery processing every month, which can be used to run queries on this public dataset. Watch this short video to learn how to get started quickly using BigQuery to access public datasets. What is BigQuery .

Dataset Title: OpenStreetMap Quality Assurance Dataset

Dataset Description: This dataset comprises OpenStreetMap (OSM) data obtained from the Dublin area in 2023, specifically for quality assurance purposes. The dataset contains a diverse range of geospatial information, meticulously sourced from OSM through the Overpass API.

Data Source: The primary source of this dataset is OpenStreetMap, accessed via the Overpass API. It encompasses a wide array of geospatial features and attributes contributed by the OSM community.

Data Format: The dataset is formatted in GeoJSON, a widely used and versatile format for representing geospatial data.

Data Size: The dataset encompasses 471 individual records, collectively forming a comprehensive representation of the Dublin area within the scope of the year 2023.

Data License: The dataset is released under the Open Database License (ODbL), ensuring openness and accessibility to users while respecting OSM's data sharing principles.

Temporal and Spatial Coverage: The dataset captures geospatial information within the vibrant city of Dublin, offering a snapshot of the region during the year 2023. It provides valuable insights into the dynamic nature of the city's geographical data.

This dataset serves as a valuable resource for quality assurance and evaluation of geospatial data within the Dublin area. Researchers, GIS professionals, and the broader OSM community can utilize it for a variety of spatial analysis and data quality assessment tasks.

Data sets used in the analysis of OpenStreetMap data quality issues, as indicated by OSM Contributors themselves, using the FIXME Feature Tag. Text Mining and Knowledge Exposition was done using Topic Models and Latent Labelled Dirichlet Allocation (L-LDA) algorithm. The data sets are taken from OSM North America and Australia and the text corpus processed in a format that is suitable to be used for L-LDA.

Daylight is a complete distribution of global, open map data that’s freely available with support from community and professional mapmakers. Meta combines the work of global contributors to projects like OpenStreetMap with quality and consistency checks from Daylight mapping partners to create a free, stable, and easy-to-use street-scale global map.

The Daylight Map Distribution contains a validated subset of the OpenStreetMap database. In addition to the standard OpenStreetMap PBF format, Daylight is available in two parquet formats that are optimized for AWS Athena including geometries (Points, LineStrings, Polygons, or MultiPolygons). First, Daylight OSM Features contains the nearly 1B renderable OSM features. Second, Daylight OSM Elements contains all of OSM, including all 7B nodes without attributes, and relations that do not contain geometries, such as turn restrictions.

Daylight Earth Table is a new data schema that classifies OpenStreetMap-style tags into a 3-level ontology (theme, class, subclass) and is the result of running the earth table classification over the latest release (v1.18) of the Daylight Map Distribution. The Daylight Earth Table is available as parquet files on Amazon S3.

This dataset consists of a review of research publications about the OpenStreetMap (OSM) project published from 2016 to 2019.

The dataset was obtained as follows. First, papers published between 2016 and 2019 were extracted using Google Scholar with a query identifying all records with at least one of the keywords “OpenStreetMap” and “OSM” in the title. The extracted records were further filtered to only keep papers having a minimum length of 4 pages and published in academic journals or conference proceedings. In addition, irrelevant papers (e.g. using “OSM” as an acronym for another purpose) and non-English papers were removed from the dataset. The remaining paper were then analyzed and manually classified.

The attributes included in the dataset are the following:

• id [paper ID]
• Paper Citation [citation of the paper]
• Authors’ disciplines — multiplicity: 1-7; allowed values: computer science, informatics, social sciences, geo-information, engineering, exact sciences, interdisciplinary
• Journal’s discipline — multiplicity: 1; allowed values: computer science, informatics, social sciences, geo-information, engineering, exact sciences, interdisciplinary
• Topic(s) — multiplicity: 1:10; allowed values: application, data quality, contribution behaviours, analyzing contributions, contributors, shaping contributions, OSM effects, data enrichment, development, review
• Authors' Geography (countries) — multiplicity: 1:n; allowed values: [list of countries]
• Authors' Geography (continents) — multiplicity: 1:n; allowed values: [list of continents]
• Study Area Geography (countries) — multiplicity: 1:n; allowed values: [list of countries], [list of continents], Global, NA
• Geographic Correspondence— multiplicity: 1; allowed values: no correspondence, partial affiliation/partial location, partial affiliation/full location , full correspondence, NA
• Perspective on the community — multiplicity: 1; allowed values: OSM as a data source, data produced by contributors, a collaborative project based on contributors, contributors producing data, a unified community, a diverse community, a social product
• Evidence of engagement — multiplicity: 1-8; allowed values: none, acknowledgement, meaningful development, occasional development, object of study - direct, object of study - indirect, contribution, participation
• General comments [comments on the paper]

Here we provide the data for reproducing the analysis and figures presented in the global urban OSM building completeness analysis manuscript.Update v2024: Included OSM completeness values for 2024 and updated covariates.

List of level crossings in metropolitan France and the overseas territories.

Licence This data is derived from crowdsourcing by contributors to the OpenStreetMap (OSM) project. They are licensed under the ODbL license which requires sharing in the same way and the mandatory attribution statement must be “© OpenStreetMap contributors under the ODbL license” in accordance with the license detailed on page http://osm.org/copyright.

Use of data In OpenStreetMap grade crossings are marked ‘railway=level_crossing’ and all associated data are documented on the dedicated OpenStreetMap wiki page.

The OpenStreetMap community Openstreetmap is the “Mapping Wikipedia”, a coordinated, self-organised global community creating freely usable data. Openstreetmap is today considered to be the most comprehensive open map database in the world.

The OpenStreetMap community strives to accurately and publicly describe the mapping specifications of all data created in OpenStreetMap wiki. However, no guarantee of quality or completeness is provided. If you identify gaps or data that should be corrected, you are welcome to do so yourself.

In France, the OpenStreetMap France Association supports the community. Openstreetmap relies largely on volunteer work. If this dataset has been helpful to you, you can support the OpenStreetMap France association by donating.

OpenStreetMap Data Exports There are several other ways to download data from the OpenStreetMap database: 1. Search datasets published by OpenStreetMap on data.gouv.fr. The majority of data is updated daily directly from OSM. 2. Use overpass turbo to specify exactly the data to download, worldwide. Formats CSV, GeoJSON, GPX, KML, OSM and more. A more complete but more technical solution. The documentation is available on OpenStreetMap wiki. 3. Use another solution. Many export possibilities exist and are listed on OpenStreetMap wiki page.

Contact If you have any questions regarding OpenStreetMap data exports, you can contact OpenStreetmap France volunteers: exports@openstreetmap.fr — https://www.openstreetmap.fr — Twitter: @OSM_FR — Mastodon: @osm_fr

— Updates — 19/03/2021: first publication of the dataset. The company Mapbox, which is part of the OpenStreetMap community, has carried out in recent days a work to improve the quality of crossings in France in OpenStreetMap. This upgrade is based on data published by SNCF. Details related to this project are public: https://github.com/mapbox/mapping/issues/375

StreetSurfaceVis

StreetSurfaceVis is an image dataset containing 9,122 street-level images from Germany with labels on road surface type and quality. The CSV file streetSurfaceVis_v1_0.csv contains all image metadata and four folders contain the image files. All images are available in four different sizes, based on the image width, in 256px, 1024px, 2048px and the original size.
Folders containing the images are named according to the respective image size. Image files are named based on the mapillary_image_id.

You can find the corresponding publication here: StreetSurfaceVis: a dataset of crowdsourced street-level imagery with semi-automated annotations of road surface type and quality

Image metadata

Each CSV record contains information about one street-level image with the following attributes:

• mapillary_image_id: ID provided by Mapillary (see information below on Mapillary)
• user_id: Mapillary user ID of contributor
• user_name: Mapillary user name of contributor
• captured_at: timestamp, capture time of image
• longitude, latitude: location the image was taken at
• train: Suggestion to split train and test data. `True` for train data and `False` for test data. Test data contains data from 5 cities which are excluded in the training data.
• surface_type: Surface type of the road in the focal area (the center of the lower image half) of the image. Possible values: asphalt, concrete, paving_stones, sett, unpaved
• surface_quality: Surface quality of the road in the focal area of the image. Possible values: (1) excellent, (2) good, (3) intermediate, (4) bad, (5) very bad (see the attached Labeling Guide document for details)

Image source

Images are obtained from Mapillary, a crowd-sourcing plattform for street-level imagery. More metadata about each image can be obtained via the Mapillary API . User-generated images are shared by Mapillary under the CC-BY-SA License.

For each image, the dataset contains the mapillary_image_id and user_name.
You can access user information on the Mapillary website by https://www.mapillary.com/app/user/
and image information by https://www.mapillary.com/app/?focus=photo&pKey=

If you use the provided images, please adhere to the terms of use of Mapillary.

Instances per class

Total number of images: 9,122

For modeling, we recommend using a train-test split where the test data includes geospatially distinct areas, thereby ensuring the model's ability to generalize to unseen regions is tested. We propose five cities varying in population size and from different regions in Germany for testing - images are tagged accordingly.

Number of test images (train-test split): 776

Inter-rater-reliablility

Three annotators labeled the dataset, such that each image was annotated by one person. Annotators were encouraged to consult each other for a second opinion when uncertain.
1,800 images were annotated by all three annotators, resulting in a Krippendorff's alpha of 0.96 for surface type and 0.74 for surface quality.

Recommended image preprocessing

As the focal road located in the bottom center of the street-level image is labeled, it is recommended to crop images to their lower and middle half prior using for classification tasks.

This is an exemplary code for recommended image preprocessing in Python:

from PIL import Image
img = Image.open(image_path)
width, height = img.size
img_cropped = img.crop((0.25 * width, 0.5 * height, 0.75 * width, height))

License

CC-BY-SA

Citation

If you use this dataset, please cite as:

Kapp, A., Hoffmann, E., Weigmann, E. et al. StreetSurfaceVis: a dataset of crowdsourced street-level imagery annotated by road surface type and quality. Sci Data 12, 92 (2025). https://doi.org/10.1038/s41597-024-04295-9

@article{kapp_streetsurfacevis_2025,
title = {{StreetSurfaceVis}: a dataset of crowdsourced street-level imagery annotated by road surface type and quality},
volume = {12},
issn = {2052-4463},
url = {https://doi.org/10.1038/s41597-024-04295-9},
doi = {10.1038/s41597-024-04295-9},
pages = {92},
number = {1},
journaltitle = {Scientific Data},
shortjournal = {Scientific Data},
author = {Kapp, Alexandra and Hoffmann, Edith and Weigmann, Esther and Mihaljević, Helena},
date = {2025-01-16},
}

-----------------------------------------------------------------------------------------------------------------------------------------------------------

This is part of the SurfaceAI project at the University of Applied Sciences, HTW Berlin.

- Prof. Dr. Helena Mihajlević
- Alexandra Kapp
- Edith Hoffmann
- Esther Weigmann

Contact: surface-ai@htw-berlin.de

https://surfaceai.github.io/surfaceai/

Funding: SurfaceAI is a mFund project funded by the Federal Ministry for Digital and Transportation Germany.

This web map references the live tiled map service from the OpenStreetMap (OSM) project. OpenStreetMap (OSM) is an open collaborative project to create a free editable map of the world. Volunteers gather location data using GPS, local knowledge, and other free sources of information and upload it. The resulting free map can be viewed and downloaded from the OpenStreetMap server: http://www.OpenStreetMap.org. See that website for additional information about OpenStreetMap. It is made available as a basemap for GIS work in ESRI products under a Creative Commons Attribution-ShareAlike license. Tip: This service is one of the basemaps used in the ArcGIS.com map viewer and ArcGIS Explorer Online. Simply click one of those links to launch the interactive application of your choice, and then choose Open Street Map from the Basemap control to start using this service. You'll also find this service in the Basemap gallery in ArcGIS Explorer Desktop and ArcGIS Desktop 10. Tip: Here are some well known locations as they appear in this web map, accessed by launching the web map with a URL that contains location parameters:AthensCairoJakartaMoscowMumbaiNairobiParisRio De JaneiroShanghai

OpenStreetMap (OSM) is a collaborative project to create a free editable map of the world. Created in 2004, it was inspired by the success of Wikipedia and more than two million registered users who can add data by manual survey, GPS devices, aerial photography, and other free sources.

OSM is produced as a public good by volunteers, and there are no guarantees about data quality. OpenStreetMap® is open data, licensed under the Open Data Commons Open Database License (ODbL) by the OpenStreetMap Foundation (OSMF).

OSM represents physical features on the ground (e.g. roads or buildings) using tabs attached to its basic data structure (its nodes, ways, and relations). Each tag describes a geographic attribute of the feature being shown by the specific node, way or relation.

Nodes are one of the core elements in the OSM data model. It consists of a single point in space defined by its latitude, longitude and node id. Nodes can be used to define standalone point features.

†Proportion obtained by removing parallel pairs with one empty and one non-empty values. ‡Proportion obtained by treating pairs with one empty and one non-empty values as symmetrical examples.

Data from European cities with results of test for quality addresses data algorithm (paper ISPS IJGI).Addresses data used on this paper are available on these websites:A) OpenAddresses: https://batch.openaddresses.io/dataB) OpenStreetMap: https://wiki.openstreetmap.org/wiki/Downloading_dataC) Google Places: https://developers.google.com/maps/documentation/places/web-service/overview?D) Bing: https://learn.microsoft.com/en-us/bingmaps/rest-services/locations/E) Here: https://developer.here.com/documentation/geocoding-search-api/*NOTES: Due to rights and property reasons, we can not distribute commercial and authoritative addresses data used on this study

The data and codes that support the findings of the manuscript entitled "A comprehensive quality assessment framework for linear features from Volunteered Geographic Information".

Public urban green spaces are important for the urban quality of life. Still, comprehensive open data sets on urban green spaces are not available for most cities. As open and globally available data sets the potential of Sentinel-2 satellite imagery and OpenStreetMap (OSM) data for urban green space mapping is high but limited due to their respective uncertainties. Sentinel-2 imagery cannot distinguish public from private green spaces and its spatial resolution of 10 meters fails to capture fine-grained urban structures, while in OSM green spaces are not mapped consistently and with the same level of completeness everywhere. To address these limitations we propose to fuse these data sets under explicit consideration of their uncertainties. The Sentinel-2 derived Normalized Difference Vegetation Index was fused with OSM data using the Dempster-Shafer theory to enhance the detection of small vegetated areas. The distinction between public and private green spaces was achieved using a Bayesian hierarchical model and OSM data. The analysis was performed based on land use parcels derived from OSM data and tested for the city of Dresden, Germany. The overall accuracy of the final map of public urban green spaces was 95\%, which was mainly influenced by the uncertainty of the public accessibility model.

This dataset provides detailed information on road surfaces from OpenStreetMap (OSM) data, distinguishing between paved and unpaved surfaces across the region. This information is based on road surface prediction derived from hybrid deep learning approach. For more information on Methods, refer to the paper

Roughly 0.0497 million km of roads are mapped in OSM in this region. Based on AI-mapped estimates the share of paved and unpaved roads is approximately 0.0117 and 0.0111 (in million kms), corressponding to 23.4856% and 22.3095% respectively of the total road length in the dataset region. 0.0269 million km or 54.2049% of road surface information is missing in OSM. In order to fill this gap, Mapillary derived road surface dataset provides an additional 0.0008 million km of information (corressponding to 3.0011% of total missing information on road surface)

It is intended for use in transportation planning, infrastructure analysis, climate emissions and geographic information system (GIS) applications.

This dataset provides comprehensive information on road and urban area features, including location, surface quality, and classification metadata. This dataset includes attributes from OpenStreetMap (OSM) data, AI predictions for road surface, and urban classifications.

AI features:

• pred_class: Model-predicted class for the road surface, with values "paved" or "unpaved."

• pred_label: Binary label associated with pred_class (0 = paved, 1 = unpaved).

• osm_surface_class: Classification of the surface type from OSM, categorized as "paved" or "unpaved."

• combined_surface_osm_priority: Surface classification combining pred_label and surface(OSM) while prioritizing the OSM surface tag, classified as "paved" or "unpaved."

• combined_surface_DL_priority: Surface classification combining pred_label and surface(OSM) while prioritizing DL prediction pred_label, classified as "paved" or "unpaved."

• n_of_predictions_used: Number of predictions used for the feature length estimation.

• predicted_length: Predicted length based on the DL model’s estimations, in meters.

• DL_mean_timestamp: Mean timestamp of the predictions used, for comparison.

OSM features may have these attributes(Learn what tags mean here):

• name: Name of the feature, if available in OSM.

• name:en: Name of the feature in English, if available in OSM.

• name:* (in local language): Name of the feature in the local official language, where available.

• highway: Road classification based on OSM tags (e.g., residential, motorway, footway).

• surface: Description of the surface material of the road (e.g., asphalt, gravel, dirt).

• smoothness: Assessment of surface smoothness (e.g., excellent, good, intermediate, bad).

• width: Width of the road, where available.

• lanes: Number of lanes on the road.

• oneway: Indicates if the road is one-way (yes or no).

• bridge: Specifies if the feature is a bridge (yes or no).

• layer: Indicates the layer of the feature in cases where multiple features are stacked (e.g., bridges, tunnels).

• source: Source of the data, indicating the origin or authority of specific attributes.

Urban classification features may have these attributes:

• continent: The continent where the data point is located (e.g., Europe, Asia).

• country_iso_a2: The ISO Alpha-2 code representing the country (e.g., "US" for the United States).

• urban: Binary indicator for urban areas based on the GHSU Urban Layer 2019. (0 = rural, 1 = urban)

• urban_area: Name of the urban area or city where the data point is located.

• osm_id: Unique identifier assigned by OpenStreetMap (OSM) to each feature.

• osm_type: Type of OSM element (e.g., node, way, relation).

The data originates from OpenStreetMap (OSM) and is augmented with model predictions using images downloaded from Mapillary in combination with the GHSU Global Human Settlement Urban Layer 2019 and AFRICAPOLIS2020 urban layer.

This dataset is one of many HeiGIT exports on HDX. See the HeiGIT website for more information.

We are looking forward to hearing about your use-case! Feel free to reach out to us and tell us about your research at communications@heigit.org – we would be happy to amplify your work.

This is the repository for the results of the 'expert opinion survey on environmental modeling with InVEST, Mapbiomas, and Open Street Maps'.

Note: check the most recent version in the sidebar

Available files:

The column descriptions in the statistical table are as follows:

Prefixes:

• HABITAT: habitat suitability score
• WEIGHT: Threat weight
• MAX_DIST: Maximum distance of negative influence (impact)

Suffixes:

• mean: Average
• std: Standard deviation
• min: Minimum value
• p05: 5th percentile
• p25: 25th percentile
• p50: 50th percentile (median)
• p75: 75th percentile
• p95: 95th percentile
• max: Maximum value

These prefixes and suffixes describe various statistical measures used to analyze the environmental modeling data.

This dataset provides detailed information on road surfaces from OpenStreetMap (OSM) data, distinguishing between paved and unpaved surfaces across the region. This information is based on road surface prediction derived from hybrid deep learning approach. For more information on Methods, refer to the paper

Roughly 0.6464 million km of roads are mapped in OSM in this region. Based on AI-mapped estimates the share of paved and unpaved roads is approximately 0.1355 and 0.015 (in million kms), corressponding to 20.9638% and 2.315% respectively of the total road length in the dataset region. 0.4959 million km or 76.7212% of road surface information is missing in OSM. In order to fill this gap, Mapillary derived road surface dataset provides an additional 0.0026 million km of information (corressponding to 0.5269% of total missing information on road surface)

It is intended for use in transportation planning, infrastructure analysis, climate emissions and geographic information system (GIS) applications.

This dataset provides comprehensive information on road and urban area features, including location, surface quality, and classification metadata. This dataset includes attributes from OpenStreetMap (OSM) data, AI predictions for road surface, and urban classifications.

AI features:

• pred_class: Model-predicted class for the road surface, with values "paved" or "unpaved."

• pred_label: Binary label associated with pred_class (0 = paved, 1 = unpaved).

• osm_surface_class: Classification of the surface type from OSM, categorized as "paved" or "unpaved."

• combined_surface_osm_priority: Surface classification combining pred_label and surface(OSM) while prioritizing the OSM surface tag, classified as "paved" or "unpaved."

• combined_surface_DL_priority: Surface classification combining pred_label and surface(OSM) while prioritizing DL prediction pred_label, classified as "paved" or "unpaved."

• n_of_predictions_used: Number of predictions used for the feature length estimation.

• predicted_length: Predicted length based on the DL model’s estimations, in meters.

• DL_mean_timestamp: Mean timestamp of the predictions used, for comparison.

OSM features may have these attributes(Learn what tags mean here):

• name: Name of the feature, if available in OSM.

• name:en: Name of the feature in English, if available in OSM.

• name:* (in local language): Name of the feature in the local official language, where available.

• highway: Road classification based on OSM tags (e.g., residential, motorway, footway).

• surface: Description of the surface material of the road (e.g., asphalt, gravel, dirt).

• smoothness: Assessment of surface smoothness (e.g., excellent, good, intermediate, bad).

• width: Width of the road, where available.

• lanes: Number of lanes on the road.

• oneway: Indicates if the road is one-way (yes or no).

• bridge: Specifies if the feature is a bridge (yes or no).

• layer: Indicates the layer of the feature in cases where multiple features are stacked (e.g., bridges, tunnels).

• source: Source of the data, indicating the origin or authority of specific attributes.

Urban classification features may have these attributes:

• continent: The continent where the data point is located (e.g., Europe, Asia).

• country_iso_a2: The ISO Alpha-2 code representing the country (e.g., "US" for the United States).

• urban: Binary indicator for urban areas based on the GHSU Urban Layer 2019. (0 = rural, 1 = urban)

• urban_area: Name of the urban area or city where the data point is located.

• osm_id: Unique identifier assigned by OpenStreetMap (OSM) to each feature.

• osm_type: Type of OSM element (e.g., node, way, relation).

The data originates from OpenStreetMap (OSM) and is augmented with model predictions using images downloaded from Mapillary in combination with the GHSU Global Human Settlement Urban Layer 2019 and AFRICAPOLIS2020 urban layer.

This dataset is one of many HeiGIT exports on HDX. See the HeiGIT website for more information.

We are looking forward to hearing about your use-case! Feel free to reach out to us and tell us about your research at communications@heigit.org – we would be happy to amplify your work.

This dataset makes it possible to compare the location of bus stops provided by operators to Île-de-France Mobilités (corresponds to the stops currently published in the GTFS< /abbr>) and the Bus stops present in OpenStreetMap.

Objective

The objective of this comparison is to improve the quality of the data provided by the operators using the potential of OpenStreetMap crowdsourced data.

OpenStreetMap (OSM) data

The OpenStreetMap community maps the whole world street by street and participates in the creation of the largest geographical database under free license. This comparison is possible thanks to the work of this community which, since April 2018, has been enriching OSM data with identifiers from Ile-de-France Mobilités Stops Repository.

The Jungle Bus audit

This production follows on from the studies and analyzes carried out by Jungle Bus which aimed to assess the extent to which these two datasets could mutually enrich each other. In one year of work, the volume of the comparable sample has been multiplied by more than 4. More information: https://junglebus.io/les-donnees-de-transport-en-ile-de-france-dans- openstreetmap-largely-untapped-potential/

Warning: currently, approximately 70% of Stops in the Île-de-France Mobilités Repository find an equivalence in OpenStreetMap (only these Stops are presented in this dataset)

This dataset provides detailed information on road surfaces from OpenStreetMap (OSM) data, distinguishing between paved and unpaved surfaces across the region. This information is based on road surface prediction derived from hybrid deep learning approach. For more information on Methods, refer to the paper

Roughly 1.4803 million km of roads are mapped in OSM in this region. Based on AI-mapped estimates the share of paved and unpaved roads is approximately 0.3769 and 0.2051 (in million kms), corressponding to 25.4635% and 13.8563% respectively of the total road length in the dataset region. 0.8983 million km or 60.6802% of road surface information is missing in OSM. In order to fill this gap, Mapillary derived road surface dataset provides an additional 0.0047 million km of information (corressponding to 0.5263% of total missing information on road surface)

It is intended for use in transportation planning, infrastructure analysis, climate emissions and geographic information system (GIS) applications.

This dataset provides comprehensive information on road and urban area features, including location, surface quality, and classification metadata. This dataset includes attributes from OpenStreetMap (OSM) data, AI predictions for road surface, and urban classifications.

AI features:

• pred_class: Model-predicted class for the road surface, with values "paved" or "unpaved."

• pred_label: Binary label associated with pred_class (0 = paved, 1 = unpaved).

• osm_surface_class: Classification of the surface type from OSM, categorized as "paved" or "unpaved."

• combined_surface_osm_priority: Surface classification combining pred_label and surface(OSM) while prioritizing the OSM surface tag, classified as "paved" or "unpaved."

• combined_surface_DL_priority: Surface classification combining pred_label and surface(OSM) while prioritizing DL prediction pred_label, classified as "paved" or "unpaved."

• n_of_predictions_used: Number of predictions used for the feature length estimation.

• predicted_length: Predicted length based on the DL model’s estimations, in meters.

• DL_mean_timestamp: Mean timestamp of the predictions used, for comparison.

OSM features may have these attributes(Learn what tags mean here):

• name: Name of the feature, if available in OSM.

• name:en: Name of the feature in English, if available in OSM.

• name:* (in local language): Name of the feature in the local official language, where available.

• highway: Road classification based on OSM tags (e.g., residential, motorway, footway).

• surface: Description of the surface material of the road (e.g., asphalt, gravel, dirt).

• smoothness: Assessment of surface smoothness (e.g., excellent, good, intermediate, bad).

• width: Width of the road, where available.

• lanes: Number of lanes on the road.

• oneway: Indicates if the road is one-way (yes or no).

• bridge: Specifies if the feature is a bridge (yes or no).

• layer: Indicates the layer of the feature in cases where multiple features are stacked (e.g., bridges, tunnels).

• source: Source of the data, indicating the origin or authority of specific attributes.

Urban classification features may have these attributes:

• continent: The continent where the data point is located (e.g., Europe, Asia).

• country_iso_a2: The ISO Alpha-2 code representing the country (e.g., "US" for the United States).

• urban: Binary indicator for urban areas based on the GHSU Urban Layer 2019. (0 = rural, 1 = urban)

• urban_area: Name of the urban area or city where the data point is located.

• osm_id: Unique identifier assigned by OpenStreetMap (OSM) to each feature.

• osm_type: Type of OSM element (e.g., node, way, relation).

The data originates from OpenStreetMap (OSM) and is augmented with model predictions using images downloaded from Mapillary in combination with the GHSU Global Human Settlement Urban Layer 2019 and AFRICAPOLIS2020 urban layer.

This dataset is one of many HeiGIT exports on HDX. See the HeiGIT website for more information.

We are looking forward to hearing about your use-case! Feel free to reach out to us and tell us about your research at communications@heigit.org – we would be happy to amplify your work.

Policy-makers are looking to promote the uptake of bicycling as a healthy mode of travel that reduces the negative effects of traditional motorised transport (physical inactivity, air pollution, traffic congestion) and achieves sustainability goals. As an active form of mobility, bicycling improves physical and mental health and has long-term public health benefits. However, there are a number of barriers that prevent people from riding a bike, including fears about riding alongside motor vehicle traffic and the lack of safe and appropriate bicycling infrastructure. For the strategic installation of safer bicycling infrastructure or the improvement of existing infrastructure, rigorous evidence-informed scientific studies are necessary, which in turn rely on high-quality bicycling data, which is scarce. In this regard, one of the prerequisites is understanding the different types of bicycling infrastructure that exist in an urban area and create an inventory dataset that can form the basis of future bicycling-related research. OpenStreetMap (OSM) is a valuable open-source map database that contains transport infrastructure data among other things and has spatial coverage for almost the entire planet. Hence, it is used extensively by researchers and planners and it helps develop methods that are transferable and thus can be replicated irrespective of the study area. We, the Sustainable Mobility and Safety Research Group (SMSR) at Monash University, Australia, have developed a classification process to classify existing bicycling infrastructure across Greater Melbourne, Australia. We have derived knowledge from existing studies and calibrated our classification system to suit local tagging practices.