and clustering for dual polarimetric SAR data (HH-HV or, VV-VH)

This repository helps to make cluster using the target characterozation parameter, and the scattering entropy, information. The clustering plane and its feasible region are shown in the Figure 1.

Figure 1. Clustering plane using and . Shaded region in black colour is non feasible. The two bounding curves are denoted as "curve I" and "curve II".

The names and colours of zones are given in Figure 2.

Figure 2. Clustering zones and their names. It should be noted that for most of the natural targets varies from to . Hence, the clustering zones are shown within this range.

Here is an example of changes of clustering zones with time over an agricultural test site

Up and Run

Requirements:

1. Python 3
2. Gdal, numpy, matplotlib
3. Dual pol SAR data (HH--HV or, VV--VH) in C2 format

a. Run the theta_xP.py file to generate image.

b. Use PolSARpro software to compute the entropy image.

c. Rename the entropy image to H_xP.bin and H_xP.hdr.

d. Run Spatial_clustering_xP.py to generate the clustered image in png format.