Gerald Evenden gerald.evenden at verizon.net
Fri Jan 23 21:19:07 EST 2004

There are about 15 pages of heavy math in PP1395 for 
earth and circular/noncircular orbit.  To start from scratch with this 
there needs some research to be performed.  First of all, Simpson Rule
intergration goes out the window and kernel evaluation of the integrals
needs to be done.  Along with a lot of understanding of what John is
trying to say.  At the moment, each satellite appears to need some
computational intensive work to get series coefficients.  Thus, do you
want to do this for every usage or catalog new satellites to replace
(add to) the older selections?

It is beginning to look like a good project for some young stud looking
to make a good impression on his boss.

On Jan 23, 2004, at 8:45 PM, Clifford J Mugnier wrote:
> John originally published his SOM the same month I published a paper on
> single-ray rectification in "Photogrammetric Engineering."  I think it 
> was
> '77 or '78.  (You know you're getting old when you can't remember the
> year).  Anyway, I've got some of John's old typewritten and handwritten
> manuscripts as well as his hardbound USGS volume on his SOM.  No, you 
> can't
> have it.
> However, let me know what details you think you need, I'll copy or scan
> them and mail them to you.
> Cliff Mugnier
> -------------------------------------
> Oh boy!  The only amount that I have looked at it was many years ago 
> and
> that was only to check that it could replicate JPS' test numbers.  I
> will give it
> a quick look and see if I understand any of it.
> At the moment I am about half way through omerc and trying to figure 
> out
> a computation in gctp/prpj that is not documented anywhere.  BTW, EPSG
> is
> *not* a very good code source.  What is in there is basically JSP'
> workup
> but lacks some critical details.  It is not from Hotine.
> On Jan 23, 2004, at 4:45 PM, Frank Warmerdam wrote:
>> Gerald Evenden wrote:
>>> Space obl.. merc is in PROJ.4, basically a copy of gctp.
>>> +proj=lsat
>> Gerald,
>> It's embarrassing what I don't know sometimes.  I see it is driven by
>> the lsat (presumably land sat number) and path values much like the 
>> old
>> GCTP formulations I have seen.  Would it be hard to update the code to
>> support the explicit parameters JPL is suggesting?
>>  3 Inclination of orbit ascending node.  (angular units)
>>  4 Longitude of ascending orbit at equator.  (angular units)
>>  7 Period of satellite revolution.  (time in minutes)
>>  8 Satellite ratio.  This controls the start and end points of the
>> map on
>>     the earth surface.
>> My presumption is that these are being computed internally from the
>> landsat
>> number and path ... is that right?  Are these JPL guys going at the
>> problem
>> in a good way to reduce the projection to it's underlying parameters
>> or is
>> it just going to make it a huge hassle for people to enter?
>> Best regards,
> ----------------------------------------
> PROJ.4 Discussion List
> See http://www.remotesensing.org/proj for subscription, unsubscription
> and other information.
Jerry and the low riders: Daisy Mae and Joshua

PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.

More information about the Proj mailing list