[Talk-GB] Surrey air survey
John Robert Peterson
jrp.crs at gmail.com
Fri Apr 30 16:04:11 BST 2010
By the looks of the imagery edges, it's made up out a of a huge number
of small images that seem to have very odd alignments.
I would guess form my experiance that the imagery has a general error
where there is an offset that covers the whole set, or large swathes
of it, caused by GPS drift, or other similar consistant issues -- this
is (teoretcially) fairly simple to correct for, given a decent amount
of reference points.
But indevidual images likelly have an error in them too, and this
would be very tricky to correct for without taking an inordinate
number of points, and then correcting the pre stiched base images, I
don't think we have access to this, do we?
so the answer is "sort of, in theory"
in practice: if anyone has found a piece of software that will take a
large volume of image data and reproject it, please share.
On 30 April 2010 15:52, Jonathan Bennett <openstreetmap at jonno.cix.co.uk> wrote:
> On 30/04/2010 11:48, Andy Allan wrote:
>> However, the new owners of the imagery gave us it to use for tracing,
>> so I'm pretty sure any licensing issues are solvable technicalities,
>> so I'd say crack on with using it.
> That's exactly what I've been doing. However, I've noticed a varying
> amount of agreement between the imagery and traces I've taken over the
> years. Sometimes it's bang on, sometimes there's up to 5 metres of
> difference between the two. I realise this could be down to the traces
> being inaccurate, but I've seen places where the average of many tracks
> still doesn't quite match the imagery.
> If I get the chance I'll take some reference points around Guildford
> that are easily identified on the images to see how far they're out, if
> at all. If I do manage this, is there any way of using the information
> to reproject the images?
> Jonathan (Jonobennett)
> Talk-GB mailing list
> Talk-GB at openstreetmap.org
More information about the Talk-GB