Rice mapping with polarimetric SAR Data using Sentinel-1

Toolbox

Objectives:

How to process RADARSAT-2 polarimetric products;

Derive Polarimetric decompositions for rice identification;
Rice mapping using unsuperivised polarimetric classification and supervised wishart
classification.

Data:

The sample data used in this exercise was a subset from one RADARSAT-2 Fine
quad-pol image acquired on June 27, 2012.

Pass Asending

Product Single Look Complex (SLC)

Incidence angle mid swath 38.89

Resolution 8.1*7.6 (R*A)

Location Jiangsu, China

Software:

The Sentinel-1 Toolbox and SNAP, open source software for scientific learning,
research and exploitation of the large archives of ESA Sentinel and heritage missions.
Tutorial:

Open the Data:

Use the Open Product button in the top toolbar and browse for the location of

the sample data.

View the product:

In the Products View you will see the opened product. Within the product bands you
will see four polarizations:
HH
HV
VH
VV
For each polarization, there will be the complex data i and q bands and virtual bands
for intensity.
View a band:
To view the HH band, double-click on the Intensity_HH band. Zoom in using the
mouse wheel and pan by clicking and dragging the left mouse button.

Calibrating the Data

To properly work with the SAR data, the data should first be calibrated. Calibration
radiometrically corrects a SAR image so that the pixel values truly represent the radar
backscatter of the reflecting surface. Calibration is essential for quantitative use of
SAR data.
From the SAR Processing menu, go to the Radiometric menu and select Calibrate.

For polarimetric processing the data must be complex. In the Processing Parameters
tab, check mark the Save as a complex output.

Polarimetric Matrix Generation

All the polarimetric tools work with either Coherency or Covariance matrices as
input.
From the SAR Processing menu, go to the Polarimetric menu and select
Polarimetric Matrix Generation.
In the Processing Parameters tab, select a T3 matrix to convert the Quad Pol product
into a Coherency matrix T3. Press Run to begin processing.

When the processing completes, a new product will be added to the Products View.
You will notice the new bands produced correspond to the elements of the T3 matrix.
Polarimetric Speckle Filtering
To clean up some of the speckle inherent in SAR images, you can apply a speckle
filter. For full polarimetric data, speckle filters must take advantage of all bands and
preserve the complex information.
Select Polarimetric Speckle Filter from the Polarimetric menu.

In the Processing Parameters tab, select the Refined Lee speckle filter, window size
7×7. Press Run to begin processing.
Open the T11 band in both the T3 product and in new speckle filtered T3 product to
compare before and after images. The resulting image will have less speckle but also
appear more blurred.
Polarimetric Decompositions
Polarimetric decompositions allow the separation of different scattering contributions
and can be used to extract information about the scattering process.
From the SAR Processing menu, go to the Polarimetric menu and select
Polarimetric Decomposition.
In the Processing Parameters tab, select the Freeman-Durden Decomposition and
H-A-Alpha decomposition, window size are both 5×5. Press Run to begin
processing.

When the processing completes, you can view all three bands in an RGB colour view
by right-clicking on the product name and selecting Open RGB Image View from
the popup menu.
 Direct-seedling
rice field field

Transplanted
rice field

Freeman RGB（R=double bounce G=volume B=surface ）

Cloude RGB (R=entropy G=Anisotropy B=alpha)

Unsupervised polarimetric classification

From the SAR Processing menu, go to the Polarimetric menu and select
Unsupervised polarimetric classification.

In the Processing Parameters tab, select the H-Alpha Wishart and Freeman-
Durden Wishart. Press Run to begin processing.

Double-click on the class results to open it.

H-Alpha Wishart

Freeman Wishart
Supervised polarimetric classification

Select areas as training data sets using the Create a new geometry container and
other drawing tools on the right hand side of the tool box.

Create a new geometry container, name it and draw it in the image using the drawing
tools.
Select Supervised Classification Training from the Polarimetric menu, then
highlight the training geometries and click on "OK" to start the training. The center
for the coherency matrices of the pixels in each user identified class is computed and
save in a text file in user specified directory.
From the SAR Processing menu, go to the Polarimetric menu and select Supervised
Wishart Classification.

In the Processing Parameters tab, select the training geometries txt file, window size
5×5. Press Run to begin processing.
 Crops mapping with polarimetric SAR Data using GF-3 data

Objectives:

How to process GF-3 polarimetric products using PolSARpro 6.0

Polarimetric decompositions of GF-3 data for Crops identification

Crops mapping using unsuperivised polarimetric classification and

supervised wishart classification

Data: The sample data used in this exercise was a subset from one GF-3

quad-pol image acquired on July 6, 2017

Product Single Look Complex (SLC)

Incidence angle mid swath 41.18

Nominal Resolution 8

Location Guangxi, China

Center Frequency 5.40

Software: PolSARpro 6.0

Tutorial:

We choose the environment: Environment----Single Data Set (Pol-SAR),

then choosing GF-3 data file folder.

Selected and click “Save&Exit”，Will pop up one Warning：Choose “No”.

import data: import----Spaceborne Sensors----GaoFen-3(GF-3)

open file
 Read head file：Click
Import four polarizations S11\S12\S21\S22 corresponding to HH\HV\VH\VV
Extract PolSAR images, and choose Full Resolution. out data format Choose Sinclair
Element.

data conversion：Converting S2 to T3
Generate an RGB image for displaying a color composite image
Select：Display—>Create RGB File—>RGB Color Composition 1—>Run
To clean up some of the speckle inherent in SAR images, you can apply a speckle filter.
For full polarimetric data, speckle filters must take advantage of all bands and preserve
the complex information.
Next, the matrix T3 element is filtered.
Select：Process—>Polarimetric Speckle Filter—>Lee Refined Filter
Lee Refined Filter，the reason for choosing this filter is that it can retain the polarization
information well under the influence of removing speckle filters.
滤波前和滤波后的 RGB 合成图对比
Polarimetric Decompositions
Polarimetric decompositions allow the separation of different scattering contributions
and can be used to extract information about the scattering process. From Process menu,
elect Polarimetric Decomposition.
For example H/A/Alpha Polarimetric Decomposition
Unsupervised polarimetric classification
H/A/Alpha Classification:choose Process–>Polarimetric Segmentation—>H/A/Alpha
Classification
Supervised polarimetric classification
In the supervised classification method, we choose Supervised Wishart Classification
method for practice.

Click Graphic Editor, and you need to open a BMP file in order to make Classification
Training.Then, you can edit Classification Training areas.This step is very important in
supervised classification and will affect the final classification result. What’s more, you
click Run training process.
Five types of land objects were selected as follows:water, urban, forest, cropland and
bamboo forest.