Compact Polarimetric Decomposition Operator
This operator performs the following
decompositions on compact pol data:
- m-chi decomposition
- m-delta decomposition
- H-Alpha decomposition
- 2-layer RVOG decomposition
- Model-Free 3-Component Decomposition (MF3CC)
The operator first computes Stokes vector [g0, g1, g2,
g3] for each pixel from the averaged covariance matrix, then
performs user selected decomposition.
m-chi Decomposition:
where
m-delta decomposition:
where
H-Alpha decomposition:
The wave entropy is computed from the eigen decomposition of the
2x2 covariance matrix C
2
. Let λ
1
and λ
2
be the two eigen values, the entropy is given by
where
The compact pol anisotropy is defined by
RVOG decomposition:
Model-Free 3-Component Decomposition (MF3CC)
Model-free 3 component decomposition technique [4] does not
consider any volume model for the computation of the three
scattering power components. The scattering power components are
roll-invariant. The total power is conserved after decomposition
and all the scattering power components are non-negative. The
target scattering type parameter is represented as:
where, Θ
CP
is the target characterization parameter which varies from -45 to
45 degrees, m
CP
is the 2D Barakat degree of polarization, S
0
, S
1
, S
2
and S
3
are the elements of the Stokes vector and S
0
is the total power of covariance matrix C
2
.
The scattering power components are:
P
d
CP
is the even bounce power component, P
s
CP
is the odd bounce power component and P
v
CP
is the diffused power component.
Input and Output
- The input to this operator should be a simulated compact
polarimetric SAR product in complex scattering vector or complex
covariance matrix format.
- The output of this operator is the m-chi, m-delta, H-Alpha or
RVOG decomposition result represented by RGB colours.
Parameters Used
The following processing parameters are needed (see
Figure 1):
- Decomposition: m-chi, m-delta, H-Alpha, RVOG decomposition or
Model-Free 3-Component Decomposition
- Window Size X: X dimension of the sliding window used in
computing the mean covariance matrix
- Window Size Y: Y dimension of the sliding window used in
computing the mean covariance matrix
Figure 1. Dialog box for Compact
Polarimetric Decomposition operator
Reference:
[1] R. K. Raney, J. T. S. Cahill, G. W. Patterson, and B. J.
Bussey, "The m-chi decomposition of hybrid dual-polarimetric radar
data with application to lunar craters", Journal of Geophysical
Research, Vol. 117, E00H21, doi:10.1029/2011JE003986, 2012.
[2] R. K. Raney, "Hybrid-Polarity SAR Architecture", IEEE
transaction on Geoscience and Remote Sensing, Vol. 45, No. 11, Nov,
2007.
[3] S. R. Cloude, D. G. Goodenough, H. Chen, "Compact
Decomposition Theory", IEEE Geoscience and Remote Sensing Letters,
Vol. 9, No. 1, Jan. 2012.
[4] S. Dey, A. Bhattacharya, D. Ratha, D. Mandal and A. C.
Frery, 2020. “Target characterization and scattering power
decomposition for full and compact polarimetric SAR data”,
IEEE Transactions on Geoscience and Remote Sensing, 59(5),
pp.3981-3998.