|
Posted by allie on February 28, 2006, 5:20 pm
Please log in for more thread options
Jerry-
Unfortunately, I can offer no help. I was just interested in finding
out if you had figured anything out regarding this issue as I've
considered (in a very simplistic way) doing something similar.
In addition, I was wondering if you'd be willing to share how you
collected the data you do have for the photographs, especially the
coordinates for the center of the photo.
Thanks
Allie
jermuell@ufl.edu wrote:
> Dear GIS users,
>
> I recently returned from obtaining some aerial photography of the
> Nicaraguan rainforest, and have a large number of photos (.jpg's,
> from a small format camera) that I would like to geometrically correct
> and mosaic. Weather conditions were less than ideal, so despite my plan
> to take photos along nice straight E-W transects at 7000 feet, I ended
> up with photos taken at a variety of altitudes and flight directions. I
> wonder if anyone might have some advice for some of the steps that I
> would like to do.
>
> I am basically seeing it as a two-step process:
> 1. Create a photo index or uncontrolled mosaic (just to see the photo
> coverage area, and to better organize the photos).
> 2. Use the index as a departure point for working on a controlled
> mosaic.
>
> My first thought was to create a a shapefile comprised of properly
> sized, oriented, and georeferenced rectangular boxes that can be used
> to create a photo index. Does anyone know how I might generate this
> type of shapefile? The data that I have to work with are:
>
> 1. The approximate center coordinates (in dd) of each box (photo) - I
> have also created point shapefiles from these coords
> 2. The rotation parameter or bearing for each photo (in compass
> degrees), based on the flight path direction.
> 3. The approximate ground sample dimensions of a pixel in each photo
> (in m)
> 4. The approximate ground sample dimensions of each photo (in m)
>
> For creating the boxes, intermediate steps might be:
> - Find a way to calculate the corner coordinates of each photo, based
> on the center point coords, the rotation parameter and the spatial
> dimensions of each photo.
> - Enter the four corner coordinates into a text file and use this to
> generate a polygon coverage of rectangles in ArcView.
> - Any suggestions on how to do either of these steps?
>
> Another thought that I had was to create a world file (.jgw) for each
> photo, in order to correctly locate and orient it in geographic space.
> I experimented with the rotation parameters in a world file, hoping
> that I could merely enter in the bearing data for an image, and rotate
> it accordingly, but it did not work. The test image that I used did not
> respond as I had hoped. But maybe it would not be so hard to create
> world files for those photos that are not so far off from an E-W
> orientation, by just calculating the location of the upper left corner
> of the photo. Any suggestions?
>
> Thanks very much (its for a good cause),
>
> Jerry Mueller
| 
XML Sitemap