[Proj] ITRF96/UTM 36N conversion
jagoncal at gmail.com
Sun Aug 9 07:18:07 EST 2009
Mr. Gerald Evenden can clear your doubts about GEOD, since he is the author
of the program. I have used it only occasionally.
Anyway some comments on your experiments: If you are comparing distances at
millimetre level be aware that when you round your geographic coordinates
with 3 decimal places in seconds you are introducing an error of up to 1.5
cm, so your distance can be affected by a few centimetres. You can increase
the decimal places of the proj output (-W5, for 5 decimal places in seconds
or use decimal degrees).
If your distances are that small, i.e., just a few meters, then you can
calculate the distance in UTM just as the Euclidean distance
(sqrt((x2-x1)^2+(y2-y1)^2)). The deformation in UTM projection (within one
zone) is most of the times smaller than 40 cm per km, and you can calculate
it and apply it if needed.
2009/8/8 fernandohoze <muranyum at yahoo.com>
> José Alberto Gonçalves wrote:
> > In order to convert UTM-36N to long-lat with output in decimal degrees
> > can use the following PROJ command line:
> > proj -I -f "%.9f" +ellps=WGS84 +proj=utm +zone=36
> Thank you all for your replies. I tried the command José wrote (with -r
> - thanks Gerald) and it worked well. The command also works without
> reformatting the input.
> About calculating distance between two points... José, I examined the
> program you recommended but and I have some results but I need to make sure
> that they are correct. Here is what I did:
> First of all, here are the input points:
> 4415488.923061 481668.789218
> 4415490.272973 481668.977703
> I found latitude and longtitude using the following command:
> proj -r -I -f "%.9f" +ellps=WGS84 +proj=utm +zone=36
> 32.785593500 39.889265833
> 32.785595667 39.889278000
> So, when I wrote these into Google Maps (in reverse order,) it correctly
> showed the points:
> 39.889265833 32.785593500
> 39.889278000 32.785595667
> Then, I executed the following geod command:
> geod +ellps=WGS84
> As input I wrote the above numbers. As output I received the latitude and
> longtitude in degree/minute/second format:
> 39d53'21.357"N 32d47'8.137"E 180d
> 39d53'21.401"N 32d47'8.144"E 180d
> Then, I executed this command for distance calculation:
> geod +ellps=WGS84 -I +units=km
> And entered the above coordinates as input:
> 39d53'21.357"N 32d47'8.137"E 39d53'21.401"N 32d47'8.144"E
> The result was:
> 6d59'12.829" -173d0'47.167" 0.001
> 0.001 at the end of the result tells that the distance is around 0.001 km
> that is 1 meters. So, I tried the command with +units=m and the result was
> 6d59'12.829" -173d0'47.167" 1.368
> That is 1.368 meters, which seems to be correct. I did the same calculation
> using http://www.movable-type.co.uk/scripts/latlong.html this web page
> the result was 0.001369 km. So, I guess I used the correct commands. The
> inputs I used for the web page were like that:
> Lat 1: 39 53 21.357N
> Long 1: 32 47 8.137E
> Lat 2: 39 53 21.401N
> Long 2: 32 47 8.144E
> There is one problem here that I don't get. When I use cm or dm for +units
> option, I still get 1.368 insted of 136.8 for cm and 13.86 for dm.
> if I don't use any +units option, I still get 1.368 as the result. Is there
> something wrong with the command I use or am I missing something?
> I also tried everything above with the following inputs:
> 4415475.022910 481656.729116 918.850000
> 4415473.636002 481656.540539 918.850000
> And in the end I got 1.195 as the results, which is reasonable. The web
> gave 0.001194 km.
> I tried to write everything I did step by step. Could you please tell me if
> above calculations are correct or not. Also I would appreciate if you tell
> me a command that would do all the computation above.
> View this message in context:
> Sent from the PROJ.4 mailing list archive at Nabble.com.
> Proj mailing list
> Proj at lists.maptools.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Proj