Hi, When we get recent imagery over areas that have already been mapped with Bing imagery, very often it does not have exactly the same georeferencing, and very often we do not have any trace or not enough to fix this. Consequently, either the contributor will offset the new imagery on the vector data, or on the contrary will move the vector data. It results the map becomes less consistent it was before.
To solve this, when partnering with GISCorps to map Malakal <http://tasks.hotosm.org/job/413> in South Sudan, I had created a grid of offset correction points with the great offset_db plug-in for JOSM <http://wiki.openstreetmap.org/wiki/Imagery_Offset_Database> so that every contributor could, in two clicks, get the new imagery aligned on the Bing imagery, easily update the data and keep the data consistency. It worked really well. As we will get more and more access to recent imagery, my idea was to find a way to repeat this without relying to the previous action of a single person and spread the use of the offset_db plugin. Recently, in a workshop with various contributors in Burkina Faso, I tested a workflow involving the Tasking Manager and it worked also well. Next step is now to involve the crowd to this process. I created an OSM wikipage both in French and English (any correction by native speakers would be appreciated) : http://wiki.openstreetmap.org/wiki/OSM_Tasking_Manager/Offset_db and created a job for Bossangoa (http://tasks.hotosm.org/job/578) in Central African Republic, because our partners from the HIU kindly provided a recent NextView Imagery. Once you catch the workflow, you will see it is really quick to create offset corrections points and the outcome (the data consistency) are really substantial. Please join! Sincerely, Severin
_______________________________________________ HOT mailing list HOT@openstreetmap.org https://lists.openstreetmap.org/listinfo/hot
