I probably do it the hard way. The USDA flies summer color imagery in North Carolina every couple of years for crop verification. They will put draft versions of the imagery up in single image per county mosaics in MrSid format for download about 6-8 months before they make the final quarter quadragle geotiffs available. The single county mosaics are in UTM projection. In NC, we like utm zones so much, we have three of them. Since I prefer to work with a seamless image layer in the same projection (North Carolina State Plane) and quarter quad format, I:
1) Use nearblack on the county MrSid images and output to Imagine format. 2) Reproject the county Imagine images to NC State Plane projection 3) use gdaltindex to create a shapefile index (1) from an older statewide image layer with image overlap in North Carolina State Plane projection. 4) use gdaltindex to make a shapefile index (2) of the reprojected county Imagine format image files. 4) Run a python script which grabs the corner coordinates of each polygon from shapefile index 1 and use that to create a list of Imagine images that overlap with the polygon by comparing it to index 2. This script then recursively "cookie cuts" the section of each Imagine image that falls within the polygon and creates a tiff file , using gdaltranslate. After all of the "cookie cutting" has been done, the script merges all of the tiff files into one tiff file using gdalwarp. For most of this, I cheat by using python to pass os.system commands to use the utility binaries. I made up a powerpoint presentation on this a couple of years ago at an internal national meeting, is there a suitable wiki to upload it to ( after getting release permission and conversion to Openoffice impress format)? Doug Doug Newcomb USFWS Raleigh, NC 919-856-4520 ext. 14 doug_newc...@fws.gov --------------------------------------------------------------------------------------------------------- The opinions I express are my own and are not representative of the official policy of the U.S.Fish and Wildlife Service or Dept. of the Interior. Life is too short for undocumented, proprietary data formats. Hermann Peifer <pei...@gmx.eu> Sent by: gdal-dev-boun...@lists.osgeo.org 10/01/2010 09:21 AM To Jukka Rahkonen <jukka.rahko...@mmmtike.fi> cc gdal-dev@lists.osgeo.org Subject [gdal-dev] Re: gdal_rasterize -tr and -te On 01/10/2010 07:03, Jukka Rahkonen wrote: > Hermann Peifer<peifer<at> gmx.eu> writes: > > >> At work, we are taking this "standard grid" issue pretty serious, but >> indeed, we might be the only ones worldwide with such a business rule. > > You are not alone, we are reprojecting our rasters to standard grid because it > helps in making seamless mosaics from the reprojected images. We are widening > the target extents a little bit to all directions with a python script to suit > the grid. > So we are already 2 ;-) Actually, I tentatively thought there would be more. This was why I wrote to the mailing list in the first place. I am aware that this (non-)issue can be solved in a shell one-liner or perhaps X lines of some other script code. I just thought that if there is a general interest, it could perhaps be built into the library. Hermann _______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev