Forums » GAMMA Bugs »
Wish list and improvements
Added by Thorsten Seehaus over 9 years ago
Hi everybody,
We are a bunch of people using GAMMA. It is a good software which got its advantages and disadvantages like very other software or product.
I guess all of you discovered already some problems and limitations of GAMMA. So I guess we could collect some ideas, how GAMMA could be improved by adding or changing some small feature.
Since we are a relatively big user group we could ask GAMMA to consider our recommendations. This would help us and the GAMMA company to improve the software so it fits the user requirements. Well, we should of course state why it would be better and so on ....
My first recommendation:
Include a parameter to define the search radius for the offset tracking programs. The people from GAMMA told me that the search radium (or distance in range and azimuth) is defined by 1/3 of the tracking window size. Consequently, I need to define large tracking window sizes if I expect wide displacements of the features. But large tracking window sizes influence strongly the tracking results, especially in the area where the moving and non moving features are close (e.g. the lateral boundaries of glaciers). By defining the search radius and tracking window size the processing can be better adjusted to the study area and data (e.g. small features but long temporal base line -> wide displacement -> large search radium, small/medium tracking window size). Furthermore it could also reduce the computation time in areas where you need large tracking windows but the expected displacements are quite small.
Replies (2)
RE: Wish list and improvements - Added by Jacqueline Tema Salzer over 9 years ago
Hi Thorsten,
I was wondering if you have heard about any developments in this respect? I am running into exactly the same issue when tracking lava flows.
Thanks,
Jackie
RE: Wish list and improvements - Added by Charles Werner over 9 years ago
Hello,
I have in the past tried to have a small window in a larger search window, but
it did not work very well, but maybe I am missing something. My opinion is that
you cannot do this very efficiently,
ie take two large patches and set everything in the reference to be blank except
the small desired sub window. You probably have to work with 2 smaller windows
and allow tracking of the offsets.
Or possible get a course map with large windows and use that as the starting
point..., subtract an initial offset field and then move to smaller windows, a
kind of iterative offset measurement.
Any suggestions welcome.
Charles