The Cloude polarimetric decomposition is generating wrong results: the surface scattering component is swapped with the volume scattering component. See the plot below showing the 3-term decomposition over a forested region.
A JIRa ticket ([SITBX-913] - JIRA) has been created t track the issue. We will look into it. Thank you for reporting the issue.
Thank you!
We have reviewed the Cloude decomposition algorithm and the code, and did not find anything wrong with the implementation. We have also compared the decomposition result with that obtained using PolSARpro and they are the same.
1 Like
Thank you! I’ll try again
Dear Jun Lu,
Please have a look at the figures below.
The first figure is a true color RGB of an Ikonos image over a selected region in Africa. The red transect is to indicate where the profiles in figures 2 and 3 will be collected.
The second figure displays the same region but the van Zyl polarimetric decomposition of an ALOS PALSAR full-pol scene. The plot shows how surface scattering, volume scattering and double bounce change along the red transect.
The third figure shows the same as slide two but using the Cloude decomposition.
I just can see how two polarimetric decompositions that are based on eigenvalue/eigenvector analysis of the covariance matrix can generate such different results.
Happy to chat further if that helps
Thank you and best wishes
Joao
Thank you for your report. We will look into it.