|
-
GGDNS
-
A
Program Library for Estimating and Removing Noise
from Gravity Gradiometry Data
GGDNS
is a program libary designed to estimate and remove the
noise contaminating gravity gradiometry data. The automatic
denoising algorithm utilizes the discrete wavelet transform
based on orhtonormal, compactly supported wavelets such
as Daubechies wavelets or symmlets. It operates on data
profiles so that the noise in data can be removed before
they are gridded for subsequent processing and interpretation.
In
the automatic mode, the program estimates the noise level
in the input data, select the optimal wavelets and resolution
for applying denoising based on the preceeding estimation,
and then carry out denoising. In order to allow maximum
user flexibility, the program can also operate in manual
mode in which the user has the full control over the choice
of different parameters. This mode is useful when the user
has detailed knowledge about the nature of noise in the
data.
Although
the program has been designed for gravity gradient data,
the underlying method is general and the program can be
used to denoise other types of potential-field data.
The
following is a screen capture of the GGDNS Graphic User
interface. The panel on the right allows the user to select
the operational mode and input different parameters as needed.
The main panel displays the profile data as the denoising
algorithm is being applied. In the example shown, the jagged
green line is a set of gravity gradiometer data from the
Gulf of Mexico, while the solid yellow line shows the corresponding
denoised result
The
manual/documentation
for the GGDNS software package is available for download.
.
|
