The Telenav OSM team just released a new version of the OpenStreetCam JOSM plugin. In the last couple of months we had improved our sign detections and improved the detection map view, by displaying aggregated traffic sign detections instead of individual detections. Traffic signs are detected per OpenStreetCam photos, in dense areas the same traffic sign is detected several times. Checking several detected signs that represent the same traffic sign in real life is time-consuming and slows down the mapping process. In order to improve the mapping process, detections that represent the same traffic sign were aggregated into a single detection.
Map View Changes
Aggregated detections are displayed for high zoom levels starting with zoom level 16. Each aggregated detection is represented by a traffic sign cluster icon rotated based on the detection heading.
Aggregated detection data
An aggregated detection can be selected by a right click from the map and unselected by a double click. When an aggregated detection is selected the detection icon along with the belonging photo locations are highlighted on the map and its information is displayed in the right side OpenStreetCam Detection panel. The photo on which the detection has the best visibility is also loaded automatically.
Loading other photos that contains the detections can be done by either selecting a photo location from the map or by pressing the Next/Previous buttons from the OpenStreetCam Detection panel. For the associated shortcuts take a look at the OpenStreetCam shortcuts from JOSM Preferences.
By default, the plugin displays the photo locations of a selected cluster, but the plugin can be configured to display also the actual detections. This can be enabled from JOSM Preferences->OpenStreetCam plugin -> Aggregated detection settings.
If enabled individual detections are connected to the corresponding photo. This visualization is used mainly for debugging purposes.
Data filters
The OSC plugin data can be filtered based on various new filters. Some of these filters were already present and others were extended.
- data to display – the type of data to be displayed for high zoom levels; by default, we display OpenStreetCam photos and aggregated detections; if both detections and aggregated detections are selected then besides aggregated detections only the detections that do not belong to any aggregation are displayed
- only mine – displays OpenStreetCam data contributed by the logged in user
- not older than – filters out data based on a given timestamp
- detection filters – are applied to detection/aggregated detection data
- mode – is applicable only to individual detections; if set filters data based on detection mode; manual detections are detections that were manually marked on the OpenStreetCam photos, while automatic detections are detections recognized automatically by our platform
- edit status – is applicable only to individual detections; if set filters detections based on edit status (if the user had edited or not in OSM the detection)
- OSM comparison – filters the data based on the status; OSM comparison represents a status of a detection regarding its presence in OpenStreetMap; this filter is useful since the mapper can visualize only detections that need to added to the map
- detection type – filter data based on detection type and subtype; this filter is useful if the mapper would like to focus on mapping only a certain type of signs
Default filter settings can be reset by pressing the “Reset” button.
If you are interested in the components used for aggregating the traffic sign detections check out the following presentation: https://youtu.be/3I8lsIttqVA?t=290.
Upcoming features
The JOSM plugin is work on progress, we are working on improving the usability and plan to add new features from time to time. In the near feature, we plan to improve the detection mapping workflows.
We hope that you enjoy the new features! If you have ideas, suggestions or encounter any issue with the plugin during editing sessions please submit either to the GitHub issue page or to the Feedback forum .
Have fun improving the map by using OpenStreetCam images!
