How to accurately assign a plus code from satellite imagery

[Bjoern Hassler]

Hello all,

How can I accurately assign a plus code from satellite imagery? Satellite imagery sometimes has offsets that run to several metres (possibly 10th of meters - though I don't know off-hand what's typical). Certainly when going between different sets of imagery, objects 'move'. 

Of course, this is a generic problem, but when e.g. when tracing over images to create data for OpenStreetMap, it doesn't matter all that much, as what I'm mapping is at least topologically correct, and (until cm-GPS becomes standard on consumer devices) that works.

However, if use plus codes to generate postcodes, I don't want to have to change these as more accurate data is released. Suppose I have a rural location near the equator, say with a few 100 buildings, and I want to assign plus codes as postcodes - what's the best option? I could measure with a phone - and get a location accurate to a few metres, and check the alignment of imagery - but that's cumbersome, as I have to do this in each location I want to record (the offset on imagery changes). 

So if I am using geodata derived from satellite images - what's the best strategy to ensure that the alignment of these images with actual WGS84 is as accurate as possible?

Thanks!

Bjoern

[Bjoern Hassler]

I guess the question is kinda answered here: https://groups.google.com/forum/#!topic/open-location-code/6pWknU4VJ0s

Bjoern

[Blake Girardot]

Hi Bjoern,

I think the best way is to use the TMS overlay in JOSM or QGIS so you
can visualize what OLC would be best for a building.

Just add this as a TMS layer in either tool:
https://grid.plus.codes/grid/tms/{z}/{x}/{y}.png and then zoom in to
the building level.

I also suggest using the full 10 digit codes only, not the shortened forms.

Let me know if I can help with any of that.

Cheers,
Blake

> --
> Public site: http://www.openlocationcode.com/
> Github project: https://github.com/google/open-location-code
> Demo site: http://plus.codes/
> ---
> You received this message because you are subscribed to the Google Groups "Plus Codes Community Forum" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to open-location-c...@googlegroups.com.
> To post to this group, send email to open-loca...@googlegroups.com.
> Visit this group at https://groups.google.com/group/open-location-code.
> To view this discussion on the web, visit https://groups.google.com/d/msgid/open-location-code/02cc2d57-c16a-4476-9ee1-951e20fc3311%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.



--
----------------------------------------------------
Blake Girardot
OSM Wiki - https://wiki.openstreetmap.org/wiki/User:Bgirardot
HOTOSM Member - https://hotosm.org/users/blake_girardot
skype: jblakegirardot

[Bjoern Hassler]

Hi Blake,

Thanks, that's helpful. I suppose my question was about knowing whether the imagery is well aligned. If I'm going to generate a fixed address from it, then it should be accurate, at least at that moment in time. (That's different from OSM features, that can get realigned over time as bertery o.agery and/or more accurate consumer 

GPS becomes available.)

Doug's post however pointed out that due to continental drift, lat/lon will change over time, though that is a more predictable change, rather than depending on inaccuracies of imagery.

Best wishes, from my mobile,
Bjoern

You received this message because you are subscribed to a topic in the Google Groups "Plus Codes Community Forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-location-code/ufflVoQbzqI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to open-location-c...@googlegroups.com.
To post to this group, send an email to open-loca...@googlegroups.com.

Visit this group at https://groups.google.com/group/open-location-code.

To view this discussion on the web, visit https://groups.google.com/d/msgid/open-location-code/CABmB%2B%2BQwEqAg_uTUGz6Y5ewr54Xu7jW27Z88kqNbfXK1OhFxOA%40mail.gmail.com.

[Blake Girardot]

Hi Bjoern,

I see.

So, clearly, using GPS tracks to align imagery before using the grid
to assign OLCs. Let me know if you need any pointers to using the gps
data in OSM or any of the GPS tms layers..

But really you are never going to get better than 1 - 3 meter
alignment with sat imagery I do not think, and consumer GPS on the
ground is even less accurate, so I find sat imagery, well aligned to
be probably the best compromise.

Luckily, since OLCs are spatially related, and 13 m2, I think it
handles the built in error bars in our tools pretty well.

Cheers,
Blake