Announcement

Collapse
No announcement yet.

EXIF GPS information Geotagging

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #16
    The reason I do not think it is the fault of GeoHack is that if I put the URL "http://toolserver.org/~geohack/geohack.php?params=39_52.45_0_N_75_34.04_0_W" into this link Geohack does the correct conversion to degrees, minutes and seconds and shows it in the correct position on the maps (as nearly as you are going to get it).

    Comment


      #17
      There is a sentence in one of the links referenced above on GeoHack that reads: "OS Grid ref: in the form AB123456 (for example). Degree of accuracy is flexible, but for most purposes a six figure reference is good, identifying a 100m square. 8 figures identifies a 10 metre square." Irfanview displays 6 decimal places. I don't know when it does the calculation or the actual number that is being fed to GeoHack. It seems there are parameters that can be set. Whether in Irfanview's core or in GeoHack is beyond me. This may explain the discrepancy of location between Google Earth and GeoHack.

      39° 52.45' could be: 39.874167 or 39.8741667 (8 decimal places) so perhaps that accounts for my confusion.
      Last edited by mrag; 17.05.2012, 07:05 PM.

      Comment


        #18
        EXIF geotagging - GPS users please try this out

        I have discussed with Irfan Skiljan the problems that have been revealed in this thread when using the GeoHack option with Irfanview. There are a number of different formats in which GPS coordinates might have been recorded in the EXIF data by cameras or other devices and it appears that a lot of work has been put into GeoHack to handle them all correctly. So it would seem to be a mistake to try to process them into a common format within Irfanview before passing them on to GeoHack.

        Irfan has produced a revised version of the Metadata plugin and sent me a copy to test. I tried the files that mrag has posted in the GPS format from his IPad camera with both Google Earth and GeoHack. I also tried GPS data from my own Panasonic camera. In all cases the results now appear to show the same locations in GeoHack as in Google Earth. In some cases those locations are not precisely where I expected them to be but I believe such errors are due to inaccuracies in the GPS values as recorded by the cameras and not how Irfanview or the other software handles the data.

        Irfan and I can only test a small sample of the GPS data produced by different cameras though, so I asked that he allow the new Metadata plugin to be posted here to let other users who work with GPS try it out with their own data before it gets issued in the next Irfanview update. So here it is. Just unzip the file and replace the current version of Metadata.dll in your plugins folder with this one. If you think that there are still errors in the locations shown for your data, please post details here so that we can look at them. In most cases if the locations have been shown in the correct locations previously I would not expect you to see any change but let us know if you do.

        Please note that this is not an invitation to request new EXIF GPS features. If you want those please discuss them in the Feature requests thread in the usual way.
        Attached Files

        Comment


          #19
          I have a similar problem with my camera. I'm using a Samsung Galaxy S2 phone. (For further details on my problem, please see this post). I've tested the new "Metadata.dll" library with almost the same results as the original.

          I think I have a theory about this problem. Probably some camera manufacturers encode the GPS latitude/longitude data using degrees only, or degrees and minutes, instead of using the three fields. So, if they use degrees only, they keep the fine accuracy on the decimals of degrees. If they use degrees and minutes, they keep the fine accuracy on the decimals of minutes.

          In this cases, if IrfanView only takes two decimals from the least significant field, the accuracy is compromised. I think it's valid to take two decimals from the seconds field, but not from degrees or minutes when the least significant fields values are zero.

          Could it be this way?

          Greetings,
          Logix.
          Last edited by logix; 14.03.2013, 01:08 AM.

          Comment

          Working...
          X