Actually, just an hour ago I figured out that my problem was in how I
stitched together the USGS tiles.  By rotating each USGS tile 2 \degr
CW everything is now lining up.  :)  However, I'm still wondering if
this "fix" will break in other regions of the country, such as at the
opposite corner of the same UTM zone.

Here's an example of the new tiles lining up perfectly:
http://wtisgpub.coe.montana.edu/raw.png

So Mapnik was perfect the whole time.  Once I'm done with this
project, hopefully I'll have some time to write up exactly what I did
to combine them together.  All USGS DOQQ tiles are public domain,
which means they might be a good start to including satellite data in
OSM.

j

On Wed, May 21, 2008 at 12:16 AM, Martijn van Oosterhout
<[EMAIL PROTECTED]> wrote:
> On Tue, May 20, 2008 at 11:40 PM, Jeffrey Sharkey <[EMAIL PROTECTED]> wrote:
>> I'm stitching together the DOQQ satellite tiles gathered using the
>> USGS Seamless service, which have a 3.33% overlap.  The DOQQ metadata
>> claims to be using NAD83 and GRS80 in UTM Zone 12.  After stitching,
>> the satellite tiles look perfect.
>>
>> When I use Mapnik+OSM data to render tiles, they don't seem to line up
>> with any consistency when overlaid with the DOQQ images.  Here is an
>> example of what I've come up with so far, where red outlines are
>> overlapping USGS tile boundaries, and green outlines are the Mapnik
>> tiles I generated.  Also included is the Python code I'm using to
>> stitch everything together:
>
> Two possibilities:
> - The OSM data comes from TIGER which is not always equally accurate.
> I would suggest first overlaying the OSM over yahoo maps using
> potlatch, that will tell you the quality of the OSM data.
> - Did you reproject the OSM data to match your output projection? I
> don't see any projections in your code, but perhaps you did it in the
> config file?
>
> Have a nice day,
> --
> Martijn van Oosterhout <[EMAIL PROTECTED]> http://svana.org/kleptog/
>



-- 
Jeffrey Sharkey
[EMAIL PROTECTED]
http://www.jsharkey.org/
_______________________________________________
Mapnik-users mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/mapnik-users

Reply via email to