Loading ...
Sorry, an error occurred while loading the content.

Re: OZF2 TO IMAGE CONVERSION

Expand Messages
  • explorerozi
    Hello - no, dont think this is issue. Keep in mind there is a difference in tfw output from old version of OE to latest version when using same image/map file
    Message 1 of 12 , Dec 1, 2007
    • 0 Attachment
      Hello - no, dont think this is issue. Keep in mind there is a
      difference in tfw output from old version of OE to latest version
      when using same image/map file (read my previous post)...can you
      explain this?

      Images I use (airphotos) are all supplied rectifed and either come
      with UTM calibration (ecw files from Langate - Western Australian
      Govt Department) or I calibrate myself in UTM i.e. calpoints are not
      specified in lat/long but in spec'd projected grid coordinates for
      that image, in my case MGA Zone 50 (most of time). So, based on my
      experience, as mentioned before, entering cal points in grid
      coordinates, and then exporting tfw file no longer works directly in
      Arcview - did before.

      Cheers
      Greg

      --- In OziUsers-L@yahoogroups.com, "rwcx183" <lgalvin@...> wrote:

      > The specific issue is, that for .map files that are not actually
      using
      > lat/lon projection, if the cal points are spec'd in lat/lon, then
      tfw
      > export results in erroneous data. This means that any .map file
      that
      > is not referring to a lat/lon image, must have the cal points
      spec'd in
      > the projected grid coordinates for that image. In the case of a
      > manually calibrated image, if you entered the coordinates in
      lat/lon,
      > then you will not be able to export a useful .tfw file. If on the
      > other hand, you entered the cal points in grid coordinates,
      then .tfw
      > file export will work as expected. Since Ozi supports only
      transverse
      > mercator grids when manually calibrating, that means that only
      > transverse mercator projected images can use the export tfw file.
      For
      > other projections, whether they're manually calibrated or imported
      > directly, the cal points will be in lat/lon and tfw file export
      will
      > not work correctly.
      >
      > J.G.
      >
    • rwcx183
      ... not ... my ... in ... I ve done the testing. If the calibration points specified in the .map file are in lat/lon and the image is using non-lat/lon
      Message 2 of 12 , Dec 1, 2007
      • 0 Attachment
        --- In OziUsers-L@yahoogroups.com, "explorerozi" <the_explorer@...>
        wrote:
        >
        > Hello - no, dont think this is issue. Keep in mind there is a
        > difference in tfw output from old version of OE to latest version
        > when using same image/map file (read my previous post)...can you
        > explain this?
        >
        > Images I use (airphotos) are all supplied rectifed and either come
        > with UTM calibration (ecw files from Langate - Western Australian
        > Govt Department) or I calibrate myself in UTM i.e. calpoints are
        not
        > specified in lat/long but in spec'd projected grid coordinates for
        > that image, in my case MGA Zone 50 (most of time). So, based on
        my
        > experience, as mentioned before, entering cal points in grid
        > coordinates, and then exporting tfw file no longer works directly
        in
        > Arcview - did before.
        >
        > Cheers
        > Greg
        >

        I've done the testing. If the calibration points specified in
        the .map file are in lat/lon and the image is using non-lat/lon
        projection, then tfw export results in erroneous data. Now, as to
        the circumstances that result in lat/lon cal point coordinates, any
        non-transverse mercator calibrated/imported image, will result in
        lat/lon cal point coordinates. If the image is calibrated manually
        using transverse mercator grid coordinates for the cal points, or it
        is imported directly and is using transverse mercator, then the cal
        points in the .map file will be in grid coordinates and tfw export
        works correctly. If you have an example image file that doesn't
        agree, by all means, post it somewhere, so we can all see what is
        the issue.

        J.G.

        P.S. You'd think that such a ridiculously over-priced, bloated piece
        of software like ArcView could read an ECW image directly without
        need for a world file, since usually the georef data is embedded in
        ECW images.
      • explorerozi
        Hello J.G. sorry wasnt disputing what you were saying was wrong - just that it doesnt apply to problem at hand. As mentioned all maps I have issue with are
        Message 3 of 12 , Dec 2, 2007
        • 0 Attachment
          Hello J.G. sorry wasnt disputing what you were saying was wrong -
          just that it doesnt apply to problem at hand. As mentioned all maps I
          have issue with are rectified and calibrated in UTM coords using UTM
          map projection...so I am not using lat/long coords and a non lat/long
          projection but UTM coords and UTM map projection. As mentioned
          before, old verion of OziEexplorer creates perfect world file while
          new version doesnt. Maybe I need to specifiy something in Arcview but
          doesnt seem to be case.

          Arcview does import ecw images (with correct header file)at click of
          one button, perfectly (which doesnt justify price but at least it has
          this fucntion). I just used this example because if you import the
          ecw image (a UTM calibrated one) into latest version of OziExplorer
          and then created worldfile it would not match coord data data
          contained in ecw header file. I dont actually use tfw files for ecw
          images...just for other format (tif,jpg etc) map images I create
          myself from screen captures that I then want to get into Arcview.

          I will try and find/create small image for you to play with - away
          from home at moment on dial up so maybe not best time.

          Thanks for your interest

          Reagrds
          Greg


          --- In OziUsers-L@yahoogroups.com, "rwcx183" <lgalvin@...> wrote:
          >
          > I've done the testing. If the calibration points specified in
          > the .map file are in lat/lon and the image is using non-lat/lon
          > projection, then tfw export results in erroneous data. Now, as to
          > the circumstances that result in lat/lon cal point coordinates, any
          > non-transverse mercator calibrated/imported image, will result in
          > lat/lon cal point coordinates. If the image is calibrated manually
          > using transverse mercator grid coordinates for the cal points, or
          it
          > is imported directly and is using transverse mercator, then the cal
          > points in the .map file will be in grid coordinates and tfw export
          > works correctly. If you have an example image file that doesn't
          > agree, by all means, post it somewhere, so we can all see what is
          > the issue.
          >
          > J.G.
          >
          > P.S. You'd think that such a ridiculously over-priced, bloated
          piece
          > of software like ArcView could read an ECW image directly without
          > need for a world file, since usually the georef data is embedded in
          > ECW images.
          >
        • rwcx183
          ... I ... UTM ... lat/long ... but ... of ... has ... Greg, I did some more experimentation and what I found, is that for images specifying UTM, Ozi does
          Message 4 of 12 , Dec 2, 2007
          • 0 Attachment
            --- In OziUsers-L@yahoogroups.com, "explorerozi" <the_explorer@...>
            wrote:
            >
            > Hello J.G. sorry wasnt disputing what you were saying was wrong -
            > just that it doesnt apply to problem at hand. As mentioned all maps
            I
            > have issue with are rectified and calibrated in UTM coords using
            UTM
            > map projection...so I am not using lat/long coords and a non
            lat/long
            > projection but UTM coords and UTM map projection. As mentioned
            > before, old verion of OziEexplorer creates perfect world file while
            > new version doesnt. Maybe I need to specifiy something in Arcview
            but
            > doesnt seem to be case.
            >
            > Arcview does import ecw images (with correct header file)at click
            of
            > one button, perfectly (which doesnt justify price but at least it
            has
            > this fucntion). I just used this example because if you import the
            > ecw image (a UTM calibrated one) into latest version of OziExplorer
            > and then created worldfile it would not match coord data data
            > contained in ecw header file. I dont actually use tfw files for ecw
            > images...just for other format (tif,jpg etc) map images I create
            > myself from screen captures that I then want to get into Arcview.
            >
            > I will try and find/create small image for you to play with - away
            > from home at moment on dial up so maybe not best time.
            >
            > Thanks for your interest
            >
            > Reagrds
            > Greg

            Greg,

            I did some more experimentation and what I found, is that for images
            specifying UTM, Ozi does indeed write erroneous data to exported .tfw
            files. The same exact image specifying a generic transverse mercator
            projection with parameters identical to those for the UTM zone,
            results in a correct .tfw file exported. I'm not sure what to make
            of that. Normally for UTM imported images, Ozi does not actually
            write the projection parameters to the Projection Setup line in
            the .map file, but rather some parameters belonging to the last non-
            UTM image. Apparently if the .map file says that the image is UTM,
            then the Projection Setup line is ignored. I thought that Ozi might
            be using the Projection Setup line by mistake during .tfw export,
            even when the image was UTM, but forcing the correct Projection Setup
            parameters, still did not produce a correct .tfw export for UTM
            images.

            I still stand by what I've said about non-transverse mercator
            projections not working for .tfw file export under any circumstances,
            due to the cal points in the .map file being in lat/lon coordinates.

            So, to summarize here, for .tfw file export to work, the following
            must be true:

            1. The cal points in the .map file must be in grid coordinates, not
            lat/lon.

            2. The .map file must specify a generic transverse mercator
            projection and not UTM.

            Perhaps some of the other grids that Ozi supports would also work?

            J.G.
          Your message has been successfully submitted and would be delivered to recipients shortly.