[Talk-GB] Georeferencing / zeroing imagery

[Silent Spike]

Take my answers with a grain of salt. I'm no expert, but these are from my
experience and current understanding.

On Sat, Sep 14, 2019 at 3:58 PM Edward Bainton <bainton.ete at gmail.com>
wrote:

> A few questions:
> - Over how wide an area does an offset obtained that way hold good?
>

This varies, areas with large elevation differences (for example) are much
harder to truly align because the offset tends to have a much higher
variance. Even outside of factors like elevation, sometimes imagery is
consistently misaligned in one region, but inconsistently misaligned in
another. I think the key thing is to just keep checking features against
GPS; other imagery layers; and other features you know to be accurate.


> - Are the old OS maps better/worse/same as this system? Are they an
> alternative for zeroing imagery?
>

Again, no hard answer here. If there is a lot of GPS data available then I
believe it's likely more accurate to use the average of that. However, in
my experience the OS maps are at least consistently accurate enough that I
would trust them if I didn't trust the GPS data in an area.


> - If I know the grid reference of somewhere (eg, Environment Agency puts a
> 10-fig reference on a plaque on their assets) is that any help?
>

Yes, this is one method of aligning a feature. JOSM editor (for example)
allows you to create a node at specific coordinates and iD has a
hotkey-openable pane which will show coordinates at the current cursor
position. Keep in mind that the source of a feature's coordinates may not
be 100% accurate, there's always a level of uncertainty and it's easier to
decide the true position if that's known too. I believe there are also
intricacies to be aware of when converting coordinate systems, but that's a
little beyond my current knowledge.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-gb/attachments/20190914/b3aabcec/attachment.html>