Getting Started with GRUS-1 Satellite Imagery

Learn how to visualize and manage GRUS-1 Satellite Imagery by utilizing Free and Open Source GIS  (Geographic Information System) software

This section contains basic information specification of GRUS-1 imagery product

1.1 Table of Image Specification. The following table includes the product summary and specification

This section contains a brief installation instruction using QGIS, a free and open source software, to perform basic satellite imagery processing

2.1. Navigate to the QGIS site by accessing this link https://qgis.org/en/site/. You have the option to select the language of your choice from the upper right hand  portion of the screen.

2.2. On the lower portion of the main page, click on the Download button. This will take you to the download link https://qgis.org/en/site/forusers/download.html.

2.3. There are a couple of download choices depending on the Operating System (OS) of the user – Windows, macOS, Linux, or BSD

2.4. For this example, we will choose Windows, 64-bit (this will also depend on the version of Windows you are using). Although new releases can be downloaded from the site, we recommend using the most stable version of the software to avoid unexpected bugs while using it. Click the link and save to your local computer folder.

2.5. Navigate to the folder where QGIS was stored and double-click the file. This will prompt the program set up. Click Next and when the user license agreement pops up, click I agree.

2.6. Click Finish once the installation has been completed.

This section includes basic methods for handling data including importing data, area clipping, and tile mosaicing.

3.1.  Downloading and Importing Data.

For this exercise, we will be using dataset from Demo Brazil. Four tiles are selected (Cell ID: S21066451, S21066450, S21056451, and S21056450) under the October 1 date. All in all, there should be 4 selected and downloaded GRUS Tiles.

Open QGIS and under the Layer tab, select Add Layer > Add Raster Layer. Alternatively, you can use Ctrl + Shift +R. This would automatically open up the Raster Tab inside the Data Source Manager dialogue box. Click the ellipsis button to upload the images you want to import in QGIS. Once the raster images have been selected, click Add button. The importing is completed once the images have appeared in the Layer Panel.

3.2. Data Tile Mosaicing. This function enables the user to merge two or more tiles together. To start with this, navigate to Raster > Miscellaneous > Merge under the Menu bar.

Once the Merge dialogue box pops up, click on the (a) Input Layers – add all the tiles you want to mosaic, (b) Output Data type – select Byte,

(c) Merged – name and save the output mosaic data with . TIF extension. Click Run and then Close once the process is finished.

Since we are using True Color Imagery product (PSM) in this example, we selected Output Data Type as Byte (8-bit). For Multispectral products (PAN/MSI), select UInt16 since it is a 16-bit imagery. To avoid confusion, always check Table 1.2 for reference.

Note: You can check the box before Open the output file after running the algorithm to immediately see the result in your existing QGIS project.

You can check the result of the tile mosaic once the processing is finished. For this manual, we named the mosaiced file as Merged as can be seen in the Layer Panel.

3.3. Area Clipping (a) Clipping Raster by Extent. In the Menu Panel, select Raster > Extraction > Clip Raster by Extent. For (a) Input layer – select the image you want an area to be clipped, (b) Clipping extent – Click on the ellipsis button and choose Select Extent on Canvas. This would prompt an area selector tool. In this case, the user can select the extent of the region bounded within a four-corner polygon. The boundary values will then be automatically filled up once the area has been selected. Under the (c) Clipped extent – save the file with the data extension .TIF. Click Run and then Close  once the process is finished. 

3.3. Area Clipping (b) Clipping Raster by Mask. Before proceeding with this process, we need to have a data mask of the region first. If a data mask is not available, users can manually create it in QGIS. We can start  by navigating to Layer > Create Layer > New Shapefile Layer.

For (a) File Name – save the shapefile layer with .SHP extension, (b) Geometry Type – select polygon, (c) Project CRS – ensure that the projection is similar to the source data. Click OK once finished.

Note: To check, you can right click the source data/raster image layer > Properties > Information > CRS. 

The created shapefile layer should appear in the Layer Panel. To start creating the boundary area, right click the shapefile layer then select Toggle Editing. This would prompt an area selector tool which allows you to select points in the region as extent of the mask. Unlike the Clip Raster by Extent tool, users can create polygons with three or more vertices.

Users can create multiple polygons in a single shapefile layer. In this example, these multiple polygons are stored in a single shapefile layer, Test Parcel. To save the file, click the icon Save Layer Edits in the Menu panel.

To proceed with the Clipping raster by mask, we will use the shapefile we have created from previous steps. Under the Raster, navigate to Extraction > Clip Raster by Mask Layer.  (a) Input Layer – select the raster you want to extract area from, (b) Mask Layer – the layer extent you want to use as clipping mask,(c) Create an output alpha band – make sure to enable this so that the output raster would generate a transparent background, (d) Clipped – name the output file with data extension .TIF. Click Run and then Close once processing is done.

Images shown below are (a)Input Layer, (b) Mask Layer, (c) Output clipped raster

This section includes different  ways to visualize data (True Color, False Color, Singleband Gray, and Singleband Pseudo color display)

4.1. True Color Composite. For the entire section of the Data Visualization tutorial, we will be using Multispectral Product as demo data. First, import the imagery into QGIS, right-click on the layer, and select Properties. In the Layer Properties dialogue box, navigate to the Symbology Tab. Under the Band Rendering, select Red Band as Band 3, Green Band as Band 2, and Blue Band as Band 1. The band assignment is similar to the Multispectral (MSI) band layer assignment as shown in the first section of this manual (Image specification)

The result of the true color composite in comparison to the default color composite is shown below:

4.2. False Color Composite. First, import the imagery into QGIS, right-click on the layer, and select Properties. In the Layer Properties dialogue box, navigate to the Symbology Tab. Under the Band Rendering, select Red Band as Band 5 Green Band as Band 3, and Blue Band as Band 2.

The result of the false-color composite in comparison to the default color composite is shown below:

4.3. Single Band Gray Visualization. Right-click on the layer, and select Properties. In the Layer Properties dialogue box, navigate to the Symbology Tab. Under the (a) Band Rendering> Render Type - select Single-band Gray. For (b) Gray Band – we select Band 5 to be visualized for this example. In (c) Color Gradient -  select Black to White. Under (d) Min/Max Value Settings, select Cumulative Count cut to eliminate very low and very high values. Click Apply and then OK.

The result of the Single-band gray visualization in comparison to the default color composite is shown below:

4.4. Single Band Pseudocolor Visualization. Right-click on the layer and select Properties. In the Layer Properties dialogue box, navigate to the Symbology Tab. Under the (a) Band Rendering> Render Type - select Single-band Pseudocolor For (b) Gray Band – we select Band 5 to be visualized for this example. Under (c) Min/Max Value Settings, select Cumulative Count cut to eliminate very low and very high values. In (d) Interpolation – select Linear, and (e) Color Ramp – for this example, we selected Viridis. Users may opt to select other color ramps that would fit their visualization application. Click Apply and then OK.

The result of the Single-band pseudocolor visualization in comparison to the default color composite is shown below:

Last updated: June 2021