| Channel | Publish Date | Thumbnail & View Count | Download Video |
|---|---|---|---|
 | Publish Date not found |  0 Views |
Mapping is time-consuming and requires a large workforce to keep maps up-to-date. Mapillary offers a different approach to collecting geospatial data using street-level imagery. This approach enables communities to collect geospatial data faster and cheaper. But can Mapillary-generated data be useful for enriching OpenStreetMap? In this study, Mapillary-extracted map data is examined against the ground truth to assess data quality for contributing to OpenStreetMap.
In this case study, positional accuracy and completeness are assessed by benchmarking Mapillary generated data against streetlight ground truth provided by Ordu Metropolitan Municipality in Altinordu district, which covers an area of 9 km2. A total of 323 streetlights were recorded in the authoritative dataset.
Completeness and positional accuracy are evaluated for each streetlight for different camera settings. To compare the impact of camera type on positional accuracy, street-level imagery was collected with three different cameras: iPhone 11, GoPro Hero 7 Black, and GoPro Max. Collected street-level imagery was uploaded to Mapillary using the Desktop Uploader. Imagery captured with different cameras is isolated during upload to compare accuracy, completeness, and correctness of map data based on camera type.
In this experiment, we validate the effectiveness of Mapillary extracted map data by focusing on streetlights by evaluating results based on completeness and positional accuracy as key performance indicators. The best result of completeness is achieved by GoPro Hero 7 Black with 87.57% in the working area and is followed by GoPro MAX with 77.30% for Mapillary extracted streetlight data. Finally, the completeness of acquired data from iPhone 11 is 71.89%.
In terms of positional accuracy, our experiment shows that street-level images captured by GoPro MAX can be extracted with a positional error of 2.02m, followed by GoPro Hero 7 Black with a positional error of 2.17m. The average positional error of street lights extracted from street-level images captured by iPhone 11 is 2.21m. This positional error is close to the precision of a single-frequency GPS receiver.
This experimental study shows that the positioning accuracy is closely related to the GPS accuracy of the capture device, and in general, a large part of the final positional error can be attributed to this. The 3D reconstruction of Mapillary can mitigate some of these effects. In addition, capturing with a large field of view has a positive effect on the accuracy. In this study, we also validate that the positional accuracy depends on several factors of the capture process; precision of the GPS receiver and also positioning hardware, resolution and quality of the images, image capture frequency of the camera, image density in the working area, and type of photo such as flat or 360.
The overall positional accuracy is below 5 m, which can be a promising solution for enriching street lighting data on OpenStreetMap and collecting street lighting inventories for municipalities and government agencies if this data is not used for surveying purposes or reference data. However, Mapillary generated data can be useful and time-saving as a supplemental data with low collection costs.
said turksever
https://2022.stateofthemap.org/sessions/DKLT7X/
#sotm2022 #Cards
Please feel free to share this video with your friends and family if you found it useful.
