I have found out that the DIgital orthos are Virginia State Plane Coordinate system.  Through the metadata on the dvd, I found that although the discs read VASPCS North, The metadata actually refers to the VASPCS south.  I metadata also refers to the units in US survey foot.  whereas the epsg file refers to metre.  I have tried using the information from the meta to create a separate projection, but it is not working correctly.  Can you see anything wrong in my translation.

Spatial_Reference_Information:
  Horizontal_Coordinate_System_Definition:
    Planar:
      Grid_Coordinate_System:
        Grid_Coordinate_System_Name: State Plane Coordinate System 1983
        State_Plane_Coordinate_System:
          SPCS_Zone_Identifier: 4502
          Lambert_Conformal_Conic:
            Standard_Parallel: 36.766667
            Standard_Parallel: 37.966667
            Longitude_of_Central_Meridian: -78.500000
            Latitude_of_Projection_Origin: 36.333333
            False_Easting: 3500000.000000
            False_Northing: 1000000.000000
      Planar_Coordinate_Information:
        Planar_Coordinate_Encoding_Method: row and column
        Coordinate_Representation:
          Abscissa_Resolution: 1.000000
          Ordinate_Resolution: 1.000000
        Planar_Distance_Units: US Survey Foot
    Geodetic_Model:
      Horizontal_Datum_Name: North American Datum of 1983 (HARN)
      Ellipsoid_Name: Geodetic Reference System 80
      Semi-major_Axis: 6378137.000000
      Denominator_of_Flattening_Ratio: 298.257222

I set up in the layer record:(here's the testing map file)

MAP
  NAME "City of Harrisonburg, Virginia"
  SIZE 400 400
  STATUS ON
# SHAPEFILE EXTEXTS
  EXTENT 1864266.6510595274 259839.01912492423 1907053.0983945231 299828.4636574
4545
#RASTER EXTENTS
#  EXTENT 11360000.0 6820000.0 11390000.0 6870000.0
  UNITS FEET
  SHAPEPATH "/opt/mapserver/www/htdocs/harrisonburg/data"
  PROJECTION
    "init=epsg:102746"
  END

  WEB
    IMAGEPATH "/opt/mapserver/www/htdocs/tmp/"
    IMAGEURL "/tmp/"
    LOG "/opt/mapserver/www/htdocs/tmp/ms.log"
  END #end web

  # 3876.sid
  LAYER
    NAME 'layer13'
    DATA '/opt/mapserver/www/htdocs/harrisonburg/data/digitalorthos/3885.sid'
    TYPE raster
    #MAXSCALE 25000
    PROJECTION
      "proj=latlong"
      "ellps=GRS80"
      "datum=NAD83"
      "lat_0=36.333333334"
      "lat_1=36.766666667"
      "lat_2=37.966666667"
      "lon_0=-78.5"
      "units=ft"
    END
    STATUS ON
  END #end layer

  # RAILROADS
  LAYER
    NAME 'Railroads'
    DATA '/opt/mapserver/www/htdocs/harrisonburg/data/railroad'
    CLASS
      STYLE
        COLOR 255 102 200
        SYMBOL 0
      END #end style
    END #end class
    TRANSPARENCY 80
    TYPE LINE
    STATUS ON
  END #end layer

END #end map


Paul A. Malabad
Director of Information Technology
City of Harrisonburg
(540) 432-7706

------ Original Message ------
From: Frank Warmerdam <[EMAIL PROTECTED]>
Date: Saturday, July 8th, 2006 10:24 AM EDT
To: Paul A Malabad <[EMAIL PROTECTED]>
Subject: Re: [UMN_MAPSERVER-USERS] raster vs shp files extents don't match

Paul A Malabad wrote:
> Corner Coordinates:
> Upper Left (11370000.000, 6860000.000)
> Lower Left (11370000.000, 6850000.000)
> Upper Right (11380000.000, 6860000.000)
> Lower Right (11380000.000, 6850000.000)
...
> Extent: (1880340.635992, 264114.932237) - (1906873.587035, 299715.644746)
...
> I have tried using the different HARN and NAD27 epsg projections in the
> layer, but I think the problem is more basic (at least I hope it is.)
> since the extents are so skewed. Any help or a pointer to the right
> manual would be greatly appreciated.

Paul,

The problem here is that without any coordinate system information for
your MrSIDs you are left playing the "guess the coordinate system" game.
And I would suggest that the rest of us out here in internet land have even
less to go on than you.

If you can't get information from the original MrSID producers, then you
will basically have to mess around with a variety of guesses. The coordinates
for your raster are enormous. If they are in meters then 11 million meters
is roughly 1/4 of the way around the world from the projection origin.

The ratio of x to y coordinates is quite different between the vector and
raster data, so if they are near each other in the real world, it is unlikely
this is just a units issue.

I hate playing guess the coordinate system.

Best regards,
--
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up | Frank Warmerdam, [EMAIL PROTECTED]
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush | President OSGF, http://osgeo.org

Reply via email to