Between the Lines in Helsingborg: Locating Assets via Imagery

Helsingborg, a forward-thinking Swedish city, makes visual geospatial data available on Mapillary for both staff and the public. Automatically extracted map features complement the city's geospatial records and help get an overview of Helsingborg's assets.

Helsingborg, a forward-thinking Swedish city, makes visual geospatial data available on Mapillary for both staff and the public. Automatically extracted map features complement the city's geospatial records and help get an overview of Helsingborg's assets.

One of the oldest settlements in Sweden, Helsingborg is home to first pedestrian street in Sweden, has a remarkable harbor and waterfront, and boasts a blend of stunning architecture from various eras. A nearly 500-year-old map appeared in the Civitates Orbis Terrarum, published in the late 1500s. In 2017, people around the world can find the latest maps of Helsingborg with a few strokes of the keyboard, including the official city map on ArcGIS Online or Eniro’s local map.

Map of Helsingborg, 1588

But moving from an early-modern atlas to the web is only the beginning. The forward-thinking municipality has taken mapping into its own hands. Like many cities across the world, Helsingborg collects and updates its own geospatial data, ensuring that both staff and citizens can access the most recent and accurate maps.

In 2013, Helsingborg captured street-level imagery of the entire municipality—an unusual and modern data collection effort compared to cartographers and urban planners in the last century. The original purpose of the imagery was monitoring ground coating and asphalt. In 2017, however, Helsingborg worked with Mapillary to expand public access to this imagery by hosting it online.

Helsingborg’s comprehensive street-level imagery is a great basis for generating map data. For example, we can extract line features from Helsingborg’s images, showing the locations of sidewalks, traffic islands, and guard rails. These, in turn, can be combined with other data sources to create a more complete view of the city’s assets. Helsingborg can particularly use Mapillary’s line features to provide an early data source for newly built areas, before they have been surveyed on the ground.

Bike lanes in Helsingborg from city data and Mapillary line features Mapillary detections can complement Helsingborg's official bike lane data

This method of generating geospatial data is a far cry from the hand-drawn map of Helsingborg from 1588. The city is well-rendered on various basemaps, and its geospatial data is currently collected mostly by human surveying and validation. Meanwhile, the street-level imagery collected by Helsingborg uses standard camera and GPS systems. All of this represents the current state of the city’s maps, but an analysis of the images presents an opportunity to fill gaps in datasets, provide updates as fresh as the imagery itself, and reduce the time and resources needed to survey the locations of the city’s assets.

In a few minutes’ processing time, Mapillary’s database produced over half a gigabyte of map data in Helsingborg. To ease the viewing process, we selected only a few parts of this data for demoing purposes and built an interactive web map allowing you to explore what we’ve detected. All of these are line features, including classes such as roads, sidewalks, and bicycle lanes.

Parking areas in Helsingborg located as line features Parking areas located across the city

Some of the more interesting line features are sometimes difficult to find in traditional city datasets, and can also be ambiguous to survey manually. We’ve detected vegetation, typically lining a road network, as well as parking areas and traffic islands. Of particular interest for transportation and road management are line features showing the locations of dashed and solid lane markings, as well as guard rails alongside roads and bridges.

Vegetation in Helsingborg located as line features Helsingborg is a green city!

We hope to continue experimenting together with Helsingborg, matching our line features against their known roads, bicycle lanes, and other infrastructure. In this demo, we used only images from Helsingborg’s official user account to build a complete picture of the city—never mind the pun—but as more recent street-level imagery appears, we can combine the sources to produce enhanced map data.

If you’d like to see a similar demonstration in your area, let us know, and we hope to put you on the modern map just like Helsingborg! Remember to check out the full demo here, and explore your own city on our website.

/Chris

Continue the conversation