Bike Ottawa members captured 450,000 images across 2,000 km in just a little over a month to gather data for bicycle advocacy work. Determining the need for infrastructure creation and maintenance, building a model for traffic stress levels, and monitoring progress are just some of the ways they are now using the imagery.
Completing the map in Ottawa with Mapillary coverage
This autumn, Mapillary teamed up with Bike Ottawa to run a #CompleteTheMap challenge focusing on Ottawa’s bicycle infrastructure. Bike Ottawa was founded over 30 years ago and is the leading advocate for bicycle safety and community efforts in the National Capital Region of Ottawa. This region includes not only the Canadian capital city, but cities such as Gatineau in Quebec. Ottawa’s bicycle infrastructure spans across several bridges, helping to connect these different communities for cyclists who live, work, and travel throughout the region.
Bike Ottawa’s motivation for collecting imagery around the city was a desire for better data that would enhance an advocacy case for infrastructure improvements. They also work closely with the local OpenStreetMap community, recognizing maps as a powerful way of presenting such data, as well as a crucial asset for any cyclist.
“We saw a lot of possibilities with how well Mapillary integrates with OSM for mapping and also lets us organize images in time and space of places that aren't otherwise covered by photography. A fantastic feature for our advocacy work, to allow people to take a virtual ride-though of places that need improvement,” commented Heather Shearer, President of Bike Ottawa.
The Alexandra Bridge connecting Ottawa and Gatineau across the Ottawa River
The Ottawa mapping challenge wrapped up with 454,934 new images, covering 2,070 kilometers of road that didn’t feature Mapillary imagery previously. “It took many of us on some strange bike rides... touring every street in a suburb, exploring every side-path, going down a swampy trail that we'd normally avoid,” said Heather. She noted that the key to achieving this result in a little over a month was teamwork and shared encouragement (while spreading the word on social media and lending out cameras likely aided, too).
In addition to this comprehensive imagery being visible on the Mapillary platform, it has also been processed to extract geospatial data indicating where different infrastructure is present in the images, such as bike lanes, bike racks, street lights, benches, and traffic signs. This provides compelling insight into the city’s urban landscape, in addition to suggesting how and where OpenStreetMap contributors can improve the map. To explore the data derived from the images, take a look at the demo of some of the automatically detected bike infrastructure or the extracted traffic signs.
Data demo of detected bike lanes and bike racks
Heather noted that Bike Ottawa lacked a clear sense of what cycling infrastructure existed and what its condition was, while the results have yielded a well-organized record: “We've now got information on whether paths are paved, how wide they are, if they need repair, and so on. In the course of discovering places to photograph, we've also been making map corrections on the fly: adding paths that were not in existing maps, and removing mapped paths that don't exist. We can also add other details, like water fountains and whether a path is lit.”
The images and traffic sign locations are available in the JOSM and iD editor tools for OpenStreetMap, while Mapillary has provided Bike Ottawa with the geospatial data for features detected in images for use in their recent mapathon. The mapathon introduced the specifically compiled OpenStreetMap Bike Ottawa Tagging Guide meant to help volunteers do a complete job of mapping the key features in OpenStreetMap, such as speed limits, street parking, bicycle paths, and road lanes.
Using the freshly captured imagery at a mapathon (photo credit: Bike Ottawa)
Once the data is complete, the plan is to use it for calculating the “levels of traffic stress”, as Heather Shearer explained. “Each road segment will be assigned a stress score, allowing us to generate a stress heat map of the city. This will show us ‘islands’ that are isolated by the high-stress roads that serve them. This will form the basis for our future advocacy work, to help better connect all parts of the city to a low-stress cycling network that's comfortable for everyone to use.”
The mapathon focused on a small area for a test case for the traffic stress map, but the mapping work will continue across the rest of the region, with a mapping task set up on the OpenStreetMap Tasking Manager that will help guide and monitor progress.
Additionally, the images and map data are useful for several other purposes. For example, many of the pathways that cyclists use are not present on any maps, so the Mapillary images combined with new additions to OpenStreetMap help to suggest and design adequate connected cycling routes where none existed previously. This is particularly relevant in the winter, when many of the cyclists’ “roads” disappear in areas that are not snowplowed.
The imagery also provides an indication of infrastructure that needs maintenance to the city government, such as dangerous roads, poor path surface, lack of signage, and obstacles to accessibility. When new image capture efforts take place in the future, the imagery provides an avenue for monitoring progress and development. While gathered under the banner of cycling advocacy, all the images are available for anyone to use for any purpose—all adding up to an immense direct and indirect impact of the Bike Ottawa community on their home region.