PCI Geomatics Help Center

How can we help you today?


PCI Geomatics -

SAR Processing In Geomatica OrthoEngine

Because of its all-weather and all-day monitoring capabilities, Radar remote sensing offers a number of advantages for Earth-surface and feature observation. Radar sensors can see through cloud and rain as well as at night, allowing for all-weather remote observation at any time. Some features, such as ice, ocean waves, and geological structures can often be seen better in radar images than in optical images.

Over the years, radar remote sensing has provided comprehensive, timely, and accurate information about the Earths surface. Due to the unique nature of radar systems, specific tools are required to extract useable information from the data. The comprehensive radar functions in Geomatica allow you to fully exploit the power of radar for use in a wide variety of applications.

The SAR Specific Model
For orthorectification of SAR data, Geomatica uses the RADAR Specific Model which uses the additional parameters in the orbit data of SAR data to reduce the amount of ground control points (GCPs) required. The extra parameters maintain the positional accuracy and high levels of detail in the model, while concurrently reducing the number of GCPs required to very few, if any at all. This math model does not use tie points since each scene is computed using the GCPs of that scene only. If you have more than eight well distributed GCPs, both Toutin's Model and the SAR Specific Model will give you similar results.

The following table describes sensor and product levels supported by the SAR Specific Model.

Sensor Supported Products
L4.1 SGP
L4.2 SCN
Envisat ASAR ASA_IMM_1P: Level 1b Image Mode Medium Resolution
ASA_IMS_1P: Level 1b Image Mode SLC
ASA_IMP_1P: Level 1b Image Mode PRI
ASA_APM_1P: Level 1b Alternating Polarization Mode Medium Resolution
ASA_APS_1P: Level 1b Alternating Polarization Mode SLC
ASA_APP_1P: Level 1b Alternating Polarization Mode PRI
ASA_WSM_1P: Level 1b Wide Swath Mode Medium Resolution
RADARSAT SLC : Single Look Complex
SGF : SAR georeferenced Fine Resolution
SGX : SAR georeferenced Extra-Fine Resolution
SSG : SAR Systematic Geocorrected
SPG : SAR Precision Geocorrected
Georeferenced for images produced in Canada PRI level for images produced by ESA
Level 1A (SCS) : Single Look Complex Slant Range
Level 1B (MGD) : Detected Ground Multi-Look

Level 1b data



RADARSAT-2 Support in Geomatica
Geomatica v10.2 supports all type of RADARSAT-2 products. Below is a table showing the RADARSAT-2 products. The bullet icon indicates all of the RADARSAT-2 products that are distributed by MDA.

Beam Mode Single Look Complex (SLC) Path Image (SGF) Path Image Plus (SGX) Map Image (SSG) Precision Map Image (SPG)
Multi-Look Fine  
ScanSAR Narrow        
ScanSAR Wide        
Extended High
Fine Quad-Pol  
Standard Quad-Pol  


The RADARSAT-2 Radar will acquire data at four different polarizations: horizontal (HH), vertical (VV) and cross (HV & VH), over a range of resolutions ranging from 100 to three meters. RADARSAT-2 will offer the following new modes:

  • selectable linear polarization
  • selectable dual polarization
  • quad polarimetric formats
  • higher resolution


The full vectorial polarization information that RADARSAT-2 retrieves enables improved imagery of the physical properties of targets illuminated by the radar. For more information on RADARSAT-2, please see http://ccrs.nrcan.gc.ca/radar/spaceborne/radarsat2/index_e.php


RADARSAT–2 Applications
There are numerous applications for the advanced capabilities of RADARSAT-2 data. Examples include the following:

  • Disaster management
  • Forestry
  • Geology
  • Hydrology
  • Marine surveillance
  • Cartography
  • Detection and classification of man-made objects
  • Oil spill detection and monitoring
  • Monitoring wetlands and snow cover
  • Ship detection
  • Sea ice discrimination
  • Agricultural applications


For more information on the RADARSAT-2 applications, please see  http://www.radarsat2.info/application/index.asp


Setting up a RADARSAT-2 Project in OrthoEngine
This example gives you the opportunity to work with Geomatica OrthoEngine to perform orthorectification of RADARSAT-2 data. Sample RADARSAT-2 data can be download from here: https://mdacorporation.com/geospatial/international/satellites/RADARSAT-2/sample-data.

Creating the Project
The first step in correcting the data is to create a project within OrthoEngine. The project file that you create contains information about the images being corrected, as well as projection information and information regarding the mathematical model used to correct the data. The three main steps in creating a project include the following:

  1. Specifying the math model;
  2. Setting the projection;
  3. Importing the imagery

The following set of instructions outline the steps for creating a project specific to working with RADARSAT-2 data. You will observe that this workflow is much the same for other supported SAR sensors such as PALSAR and ASAR.

Starting OrthoEngine on Windows Systems
In Windows, begin by starting Geomatica by clicking on Start | Program Files | PCI Geomatics | Geomatica v10.2 | OrthoEngine. Alternatively, you can click on the OrthoEngine icon on the main Geomatica toolbar. The main OrthoEngine toolbar opens.


Creating a Project and Importing Radar Data
OrthoEngine works on a project-by-project basis. Therefore, you need to open an existing project or create a new project before you gain access to OrthoEngine functions. In this lesson, you will create a new RADARSAT-2 project.

To create a new project:

  1. Click New on the File menu. The Project Information dialog opens.
  2. Click Browse. The File to Create dialog opens.
  3. Choose a directory in which to store your RADARSAT-2 OrthoEngine project.
  4. Enter the name RADARSAT-2.prj in the File name field.
  5. Click Open. The File to Create dialog closes. The path and filename appear in the File name field in the Project Information dialog.
  6. Enter the name RADARSAT2 Tutorial Project in the Name field.
  7. In the Description field, type RADARSAT2 Ortho Project.
  8. In the Math Modeling Method panel, select ‘Radar Satellite Modeling'.
  9. In the Options panel, select ‘Radar Specific Model’. Note that if you have 8 or more GCPs, you can setup your project using Toutin’s Model.
  10. Click The Project Information dialog closes. The Set Projection dialog opens.


Setting the Output Projection Parameters
The projection information needs to be set at the beginning of each project. In the Set Projection dialog, enter the projection information for the project.

  1. Select UTM from the drop-down menu in the Output Projection panel. The UTM Zones dialog opens.
  2. Click the Earth Model button and select D000 as the datum from the datum tab.
  3. Click Accept. The UTM Zones dialog opens.
  4. Click to select Zone 10 and click Accept.
  5. Click to select Northern Hemisphere.
  6. Click Accept.
  7. For the Output Pixel Spacing, type 5.0.
  8. For the Output Line Spacing, type 5.0. The value of 5.0 meters represents the desired resolution of the ortho images mosaic file.


Setting the GCP Projection

  1. Click the Set GCP Projection based on Output Projection button. The GCP Projection adopts the same settings used for the Output Projection.
  2. Click OK. The Set Projection dialog closes.

Data Input
You do not need to import RADARSAT-2 data into a PIX file using ‘Read from CD- ROM’ (or equivalent CDSAR program) as it is supported directly in our Generic Database Library (GDB) Image can be accessed directly using ‘Product.XML’ file at Ortho Generation stage.

To input the images to the current project:

The dataset below shows a RADARSAT-2 dataset that was downloaded from  http://www.radarsat2.info/product/demo_set/. The Vancouver Dataset is the one that will be used for this particular tutorial. This dataset contains all of the beam modes available from RADARSAT-2: dual polarized, high resolution, single polarized and ultra-fine mode. The file that will be used for viewing and orthorectifying in this tutorial will be product.xml. The *.tif files that are provided with the dataset are referenced by the product.xml file. The file used for this tutorial is
Vancouver_R2_Standard7_HH_HV_SGF, and can be found in the Vancouver directory at the link above.

  1. From the Processing Step menu on the OrthoEngine toolbar, select ‘Ortho Generation’.
  2. On the Ortho Generation toolbar, click Open Image button. The Open Image dialog box opens.
  3. Click New Image in the Open Image dialog. A File Selector dialog opens.
  4. Browse to the folder containing your RAW RADARSAT-2 data.
  5. Select and open product.xml file (if you cannot see the *.xml-type files, select *.xml from the “Files of Type” dropdown in the File Selector dialog).
  6. System will prompt you for calibration type: Sigma nought, Beta nought, Gamma, None. The calibration type will be listed in the Ortho Production  panel in the "uncorrected Image" section and it is the type of ortho image  that will be produced. For example, you can add the same image twice in the ortho production panel - once for beta nought, once for sigma nought and two orthos will be produced.
  7. After this step, system will prompt you to create Overviews. Click Yes.
  8. References to the imported file will appear in the Open Image dialog


Saving the Project

  1. Click File on the main OrthoEngine toolbar.
  2. Click Save in the file menu.

The project file RADARSAT2 is saved in the specified folder. In addition, OrthoEngine automatically creates a backup file every 10 minutes. The backup file uses the same file name as the project file, but with a .bk extension.

Tip; If you need to revert to the backup file, rename the backup file so that it uses the .p  rj extension. OrthoEngine can load this project file as it would any other *.prj file.

Orthorectifying RADARSAT-2 Data
When working with the Radar Specific Model, the inclusion of ground control points (GCPs) is optional. Without any GCPs, the model is calculated based on the satellite’s positioning information. Further addition of few GCPs, although not necessary, will refine the model, and as a result, improve its accuracy. This means, however, that  as long as a DEM is included in your project (either by importing or extracting a DEM from stereo radar images), the images can still be orthorectified without any GCPs.

Preparing for Orthorectification
The OrthoImage Production dialog allows you to set up and schedule the images for orthorectification. Several images can be selected and processed in batch mode.


To setup the images:

  1. On the Ortho Generation toolbar, click Schedule Ortho Generation. The Ortho Image Production dialog appears.
  2. Select image(s) from the Available Image list.
  3. Click the arrow to move all the images to the Images to Process list.
  4. By default, the ortho image will be named ofilename.pix. At this time, you can enter a different filename for the output image in the Corrected Image panel.


Selecting the Digital Elevation Model (DEM)

  1. In the DEM panel, click Browse.
  2. Locate the folder containing the DEM that you will use for Orthorectification and select it. If you do not have a DEM, you can use the demworld.pix file, located in the Geomatica etc directory.
  3. Click Open.
  4. The Database Channels dialog opens.
  5. Select the layer containing the DEM.

Note that you should specify a value for Background Elevation. If a background elevation value exists for your DEM, enter it in the area provided. If you are unaware of the background elevation value for the DEM, click the DEM Info button. The information that is reported when you click this button will be the three lowest elevation values. Typically the lowest elevation value is used as the background elevation. If you do not have a DEM, you can specify the elevation scale and offset in the DEM panel.

6. Click the DEM Info button. The DEM information will be reported as below. Enter the lowest elevation value for the Background Elevation.

7. Select the DEM in the Available Layers list, and click OK.


Orthorectifying the RADARSAT-2 Scenes

To generate the ortho images:

1. At the bottom of the Ortho Image Production dialog, click Generate Orthos.

The Ortho Production Progress Monitor opens, and shows the status of the orthorectification process for each image.

2. After all of the orthos are generated, click the Close button at the bottom of the progress monitor.

The message Ortho done will appear beside each image in the Available Images panel, indicating that the original images are now orthorectified.

Note: This is a good time to save your project file.

Viewing the orthorectified image in OrthoEngine

  1. From the OrthoEngine File menu, select Image View. A Database File Selection window opens.
  2. From the Radar_Data folder, select the output image.
  3. Click Open.

A viewer opens, displaying the orthorectified image.


Further Processing
There are a number of additional SAR processing workflows available in OrthoEngine. For the orthorectification of the RADARSAT-2 scene used above, the inclusion of additional ground control points (GCPs) could have resulted in a more accurate output. The file product.xml distributed by RADARSAT-2 already contains some  GCPs, however if the user manually collects at least eight GCPs, the GCPs included with the product.xml file will be overwritten. To create GCPs in OrthoEngine, select the GCP processing step on the OrthoEngine toolbar, and click the Collect GCPs Manually icon. Note that the collection of GCPs must be carried out after importing the imagery.


Have more questions? Submit a request


Powered by Zendesk