[OSRS-PROJ] Re: Proj.4 DLL ?

Frank Warmerdam warmerdam at pobox.com
Tue Jan 23 09:58:24 EST 2001


Bart Adriaanse wrote:
> 
> Hi Frank,
> 
> I have been spending some time reading, including the mailing list
> archive on Proj.4, i hope you will alow me to ask you two big questions
> i have.
> 
> I noticed you mentioning in a posting that the datum transformations are
> not included in the DLL and you sort of offered to help if that was
> required.
> If i look at some of the sources i get the impression it still is that
> way and i would like to estimate how hard it would be to incorporate the
> datum conversion, in the same DLL or in a seperate one. Would you think
> this to be easy to to ?

Bart, 

The datum shifting works fine in unix .so files, the only problems on
windows is that I haven't added all the entry points to the proj.def file.
It shouldn't be hard.  I would encourage you to do it, and send me back the
updated proj.def file.  If you have problems with the datum shifting support
(which is pretty new) I am glad to help. 

Mainly I think you need to add the pj_transform entry point to proj.def, but
there are a bunch of other function called less often by the application that
could also be added, such as pj_datum_transform, pj_geocentric_to_geodetic, 
pj_geodetic_to_geocentric, and pj_compare_datums. 

> Second question, have you ever heard of the dutch RD national grid
> projection, specifically if this can be handled within reasonable
> accuracy with proj.4 ?
> I have some simple fifth degree polynomal formula's from the dutch
> topographic dept. so we could even dream of adding this to Proj.4 ?
> (what i am missing is the Bessel1841->WGS84 datum shift to make it
> usable)

My lookup in the EPSG database says that the RD projection is based on
Oblique Stereographic which I currently map to the "stere" projection in
PROJ.4.  I have no personal experience with how accurate this is though. 

The EPSG database include a 7 parameter transformation to WGS84 from
Amersfoort.  You should definately check the results of this, since I
have only done decent testing on the 3 parameter datum transformations,
not the seven.  My unvalidated hand translation of EPSG CS 29882
(RD / Netherlands New) would be:

+proj=stere +lat_0=52.1561605555555 +lon_0=5.38763888888888 +k=0.9999079
            +x_0=155000 +y_0=463000 
            +towgs84=593.16,26.15,478.54,-6.3239,-0.5008,-5.5487,4.0775
            +ellps=bessel

Let me know if this produces reasonable results. 

Best regards,

---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerda
and watch the world go round - Rush    | Geospatial Programmer for Rent
----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.



More information about the Proj mailing list