While browsing OpenStreetMap around the world-famous Riviera Country Club Golf Course, I noticed that it was pretty empty.
No fairways, sand traps, greens, holes :-(
It just so happens that, a few days ago, one of our team members was at this location collecting submeter data using an iPhone.
Most smartphones nowadays have a built-in GPS device that gives you approximately (at best) ~5 meter horizontal accuracy. For this particular project, we needed something much better. With a few clicks, we connected (via Bluetooth) our Android and iPhone devices with a high precision SXBlue GPS/GNSS receiver to our AmigoCloud Data Collection Application. Since we were automatically benefitting from RTK corrections, we were getting 1cm-2cm accuracy per vertex collected!
If you are interested in the actual process of setting up a data collection project (it takes less than 5 minutes and doesn't require you to install any software in your computer) you can watch the video of our webinar last week where we walked through it (skip to 13m:26s for details).
The data collection process itself involves:
- Downloading the application from the respective store (Google Play Store / iTunes).
Using the "record in background - track feature". This enables you to automatically record your path as you walk around the perimeter of the data you are trying to collect.
- Create records using those tracks to fill out attributes / geotagged photos, etc.
This is what the collected data looks like:
You really have to zoom in to appreciate the amount of vertices in each feature:
Here is a live map with some imagery as background.
But of course, our goal is to include this data in OSM!
To do this, we exported the data out to Shapefile format. We used the AmigoCloud data portal feature to expose the data out: Here is a link to that data portal. WARNING: It also includes vertices that are above the desired accuracy - a topic for another blog post!
There are many ways to get data into OSM (and doing big automated data imports should be done in conjunction with the OSM community). After asking a few questions to OSMers far more experienced than me, I chose to use JOSM (a popular option in my local OSM group). It can read shapefiles through the Open Data extension.
Here is the data opened in JOSM:
After a brief search for the community-approved OSM tags and some proper tagging of my features in JOSM to adhere to the standards, the data was ready to be uploaded to OSM.
After a few minutes, you can see how the data starts to render in the OpenStreetMap site:
Here is a link to the live data in OSM: http://www.openstreetmap.org/#map=16/34.0438/-118.5030
I really hope whoever is mapping in that area is happy with my changes!