[OSM-talk] ping OpenAerialMap

[Martijn van Exel]

Op 9 okt 2008, om 17:20 heeft Richard Fairhurst het volgende geschreven:

> Martijn van Exel wrote:
>
>> Christiano -- Thanks for that. My guess would be that using the WMS  
>> in
>> an OAM-external environment would actually violate the recommendation
>> to use the i-cubed dataset only within OAM. My take on it would be
>> that, as long as there's still uncertainty as to licensing, you  
>> should
>> use the i-cubed data only within the OAM *website*.
>>
>> Does Potlatch show it, for example?
>
> It did at one stage but sadly hasn't done for a while. Concerns were
> expressed that some OAM content had been georeferenced by use of
> Google Maps. Ed may have just permitted this by means of his latest
> blog posting. ;)

That would be very good news for Potlatch, wouldn't it? I guess OAM  
provides better / more up-to-date imagery in some areas - and we can  
easily add locally acquired imagery and use it in Potlatch immediately.
>
>
> For the i-cubed stuff, though, I _thought_ (may be wrong) that it was
> just colour-corrected Landsat and I don't see how colour correction
> would give anyone rights over tracing.


Well, i-cubed seems to exploit this derivation of Landsat7  
commercially, so there must be something non-trivial about the post- 
processing they do, I guess.

Just to clarify - I want to use the data for the Tripod project (http://tripod.shef.ac.uk/ 
) , which is about augmenting metadata in photos mainly based on  
existing location data. In this particular part of the project, we  
want to analyze the RGB histograms of the picture and compare them to  
the local aerial imagery to determine land cover class (e.g. forest,  
field, urban, beach).
In another part of the project, we might be using OSM data to  
determine other features found in a photo (POI, streets, ...).

Martijn