[Users] ridging attribute

[Geert de Bruin]

Dear all,

Matthias Imhof of Virginia Tech wrote:

> what would be most helpful was connecting and enumerating the ridges. 
> Instead of having pixels classified as -1, 0, +1; I would need all 
> ridge pixels belonging to the same reflector labelled with a common, 
> unique index. 


This would indeed be very helpful. In fact such functionality is part of 
our currect SSIS (Sequence Straigraphic Interpretation System) 
development project. The numbering is to be used for flattening an 
entire seismic volume (3D Wheeler transform), which enables us to study 
the data in the (chrono-)stratigraphic domain. SSIS will be a commercial 
plugin to OpendTect. We are aiming for a release Q106. Academic users, 
with whom we have a signed R&D agreement will get free access to this 
plugin.

Best regards,

Geert.


>
>
> ------------------------------------------------------------------------
>
> Subject:
> Re: [Users] ridging attribute
> From:
> Matthias Georg Imhof <mgi at vt.edu>
> Date:
> Thu, 07 Jul 2005 10:26:47 -0400
> To:
> Geert de Bruin <geert.debruin at dgb-group.com>
>
> To:
> Geert de Bruin <geert.debruin at dgb-group.com>
>
>
> Geert,
>
> what would be most helpful was connecting and enumerating the ridges. 
> Instead of having pixels classified as -1, 0, +1; I would need all 
> ridge pixels belonging to the same reflector labelled with a common, 
> unique index.
>
> Matthias
>
> Geert de Bruin wrote:
>
>> Dear all,
>>
>> At the moment I am working on sequence stratigraphy (SSIS) and in 
>> order to detect particular features I constructed a simple ridging 
>> attribute (see example 
>> ftp://dgb-group.com/pub/tmp/Ridging_example.png). This attribute was 
>> originally a pre-calculation for a different attribute, but striking 
>> was the fact that it made the interpretation of structures easier, 
>> since I was not biased by the energy of events.
>> The attribute set is quite simple. Each sample is compared with one 
>> sample above and one sample below it. If it has a higher value than 
>> the sample below and above it, the sample is located on or near a 
>> positive peak. This sample is given a 1 as output, all others are 
>> made zero. The same can be done for negative peaks, consequently the 
>> sample must be smaller than the sample above and below it. It is 
>> possible to display only the positive or negative ridges, or to 
>> combine them into a new attribute.
>> See for ftp://dgb-group.com/pub/tmp/Ridging%20attribute.ppt complete 
>> explanation.
>>
>> Best regards,
>>
>> G. de Bruin
>>
>
>
-- 

-- Geert de Bruin
-- Geoscientist
-- dGB Earth Sciences B.V.
-- Nijverheidstraat 11-2, 7511 JM Enschede, The Netherlands
-- mailto geert.debruin at dgb-group.com, http://www.dgb-group.com
-- Tel: +31 53 4315155, Fax: +31 53 4315104