[OpendTect_Users] OpendTect 2D horizon picking problem

[Bert Bril]

Hi,


Guillaume wrote:

> I'm working on a research project and i am using OpendTect. I
> attached this mail a .pdf illustrating my problem with 2D horizon
> picking. I just try to pick seeds and draw 2D horizon on my seismic
> line but i can't draw a complete horizon, OpendTect doesn't allow me
> to continue... I am working on the 4.3 OpendTect version (i first
> downloaded the 4.2 version but OpendTect 4.2 needed like 5 min to
> pick 1 seed before to crash).

Let's first establish the real nature of the problem. I understand that 
you tried to pick a horizon a couple of times, and that each time you 
could not go further than a certain point. OpendTect seems to connect 
the last successful pick with a point on the left side. I presume the 
straight line is at the depth of your last pick. The point where this 
happens is always the same. Right?

My guess is that the problem is caused by a problem in the line data. 
More specifically, I think it may be a problem with the trace numbering.

If you imported from SEG-Y, then can you take a look again with the 4.3 
version. In the examine window you can see direct plots of all the 
header values for the traces in the file. Good numbering causes there to 
be just one line with 2 points annotated: the start and stop.

You can also check in Survey-Manage-2D Geometry. Or just in the 3D 
viewer, by going over the line with your mouse. Do you see any 
irregularity in the trace number of coordinate values?


Please tell me what you get, before we have to go deeper into the 
details of the data. The data itself can have a problem, or there may be 
something in there that triggers a bug. In both cases much more 
information is required to find a solution.

In all cases, I'm not sure how interesting the rest of the conversation 
would be for the mailing list. Therefore, communication with 
support at opendtect.org may be a lot better.



Bert Bril

-- 
-- Bert Bril / OpendTect developer at dGB
-- mailto:Bert.Bril at opendtect.org , http://opendtect.org