Clip Sentinel-2 images - Revision history

Jgrocha at 10:09, 7 June 2018

2018-06-07T10:09:36Z

Jgrocha: /* Clip Band sets */

2018-06-07T10:00:22Z

Clip Band sets

Jgrocha at 09:49, 7 June 2018

2018-06-07T09:49:18Z

Jgrocha at 09:38, 7 June 2018

2018-06-07T09:38:41Z

Jgrocha at 09:26, 7 June 2018

2018-06-07T09:26:44Z

Jgrocha: /* Clipping Sentinel-2 images */

2018-06-07T09:22:40Z

Clipping Sentinel-2 images

Jgrocha: /* Create the polygon */

2018-06-06T17:58:19Z

Create the polygon

Jgrocha: Created page with "Sentinel-2 images are large and have many bands. We can minimize the processing time by restricting the images to the area we want to work on. For this exercise, we can use t..."

2018-06-06T13:48:36Z

Created page with "Sentinel-2 images are large and have many bands. We can minimize the processing time by restricting the images to the area we want to work on. For this exercise, we can use t..."

New page

Sentinel-2 images are large and have many bands. We can minimize the processing time by restricting the images to the area we want to work on.

For this exercise, we can use the GADM data. Download Denmark's data from https://gadm.org/download_country_v3.html
Or use this direct link: [https://biogeo.ucdavis.edu/data/gadm3.6/gpkg/gadm36_DNK_gpkg.zip]

== Create the polygon ==

From gadm36_DNK_2, create a unique polygon containing the 4 municipalities (kommuner):
* Hillerød,
* Helsingør,
* Gribskov,
* Fredensborg

Save the layer using EPSG:32633 as the CRS, to match the same CRS as the Sentinel-2 images we have.

[[File:Captura de ecrã de 2018-06-06 15-27-21.png]]

=== Clipping polygon ===

The resulting layer "studyarea" should look like:

[[File:Captura de ecrã de 2018-06-06 15-32-20.png]]

== Clipping Sentinel-2 images ==

Now that we have both the clipping polygon and the Sentinel-2 images stored on a folder, we clip all the images to make its processing faster.

We use the SCP plugin.

Go to: SCP → Preprocessing → Clip multiple rasters

Eventualmente, atualizar a lista de raster clicando no botão de refresh
Selecionar todos os rasters
Use shapefile for clipping (escolher a camada com o beuffer calculado previamente)

← Older revision		Revision as of 10:09, 7 June 2018
Line 112:		Line 112:

	Create a new group in the layers content table, and reaarange the layers.		Create a new group in the layers content table, and reaarange the layers.
		+
		+	== Layers contents and organization ==
		+
		+	[[File:Captura de ecrã de 2018-06-07 12-08-42.png]]

← Older revision		Revision as of 10:00, 7 June 2018
Line 103:		Line 103:

	== Clip Band sets ==		== Clip Band sets ==
		+
		+	Select Band set 1
		+	Select the polygon to be used as the clipping area
		+
		+	[[File:Captura de ecrã de 2018-06-07 11-53-41.png]]
		+
		+	Create a new folder, for example Sentinel-Clip to store the new clipped images
		+
		+	Create a new group in the layers content table, and reaarange the layers.

← Older revision		Revision as of 09:49, 7 June 2018
Line 75:		Line 75:

	[[File:Captura de ecrã de 2018-06-07 11-36-48.png]]		[[File:Captura de ecrã de 2018-06-07 11-36-48.png]]
		+
		+	{\|
		+	\| Landsat 8 Bands \|\| Wavelength [micrometers] \|\| Resolution [meters]
		+	\|-
		+	\| Band 1 - Coastal aerosol \|\| 0.43 - 0.45 \|\| 30
		+	\|-
		+	\| Band 2 - Blue \|\| 0.45 - 0.51 \|\| 30
		+	\|-
		+	\| Band 3 - Green \|\| 0.53 - 0.59 \|\| 30
		+	\|-
		+	\| Band 4 - Red \|\| 0.64 - 0.67 \|\| 30
		+	\|-
		+	\| Band 5 - Near Infrared (NIR) \|\| 0.85 - 0.88 \|\| 30
		+	\|-
		+	\| Band 6 - SWIR 1 \|\| 1.57 - 1.65 \|\| 30
		+	\|-
		+	\| Band 7 - SWIR 2 \|\| 2.11 - 2.29 \|\| 30
		+	\|-
		+	\| Band 8 - Panchromatic \|\| 0.50 - 0.68 \|\| 15
		+	\|-
		+	\| Band 9 - Cirrus \|\| 1.36 - 1.38 \|\| 30
		+	\|-
		+	\| Band 10 - Thermal Infrared (TIRS) 1 \|\| 10.60 - 11.19 \|\| 100 (resampled to 30)
		+	\|-
		+	\| Band 11 - Thermal Infrared (TIRS) 2 \|\| 11.50 - 12.51 \|\| 100 (resampled to 30)
		+	\|}

	== Clip Band sets ==		== Clip Band sets ==

@@ Line 20: / Line 20: @@
 [[File:Captura de ecrã de 2018-06-06 15-32-20.png]]
-== Clipping Sentinel-2 images ==
+== Create a Band set (virtual raster) with Sentinel images ==
 Now that we have both the study area polygon and the Sentinel-2 images stored on a folder, we clip all the images to make its processing faster.
@@ Line 71: / Line 71: @@
 [[File:Captura de ecrã de 2018-06-07 10-41-56.png]]

@@ Line 32: / Line 32: @@
 [[File:Captura de ecrã de 2018-06-07 10-34-06.png]]
-The new Bnad set is added to QGIS layers.
+The new '''Band set''' is added to QGIS layers.
-We can change the style definition, to show the first 3 bands, but by a different order.
+It is added with a default style. The default style uses the first 3 bands as the RGB bands. So, band 1 is used as Red, 2 as Green and 3 as Blue.
 {|

@@ Line 22: / Line 22: @@
 == Clipping Sentinel-2 images ==
-Now that we have both the clipping polygon and the Sentinel-2 images stored on a folder, we clip all the images to make its processing faster.
+Now that we have both the study area polygon and the Sentinel-2 images stored on a folder, we clip all the images to make its processing faster.
-We use the SCP plugin.
+To clip all the images, we use the SCP plugin  → Preprocessing → Clip multiple rasters.
-Go to: SCP → Preprocessing → Clip multiple rasters
+https://media.readthedocs.org/pdf/semiautomaticclassificationmanual/latest/semiautomaticclassificationmanual.pdf#subsubsection*.284
-    Eventualmente, atualizar a lista de raster clicando no botão de refresh
+The clip tools needs a Band set to operate on. Since we will not use some bands (1, 9 and 10), we can create a band set without them.
-    Selecionar todos os rasters
-    Use shapefile for clipping (escolher a camada com o beuffer calculado previamente)
+[[File:Captura de ecrã de 2018-06-07 10-34-06.png]]

@@ Line 12: / Line 12: @@
 * Fredensborg
-Save the layer using EPSG:32633 as the CRS, to match the same CRS as the Sentinel-2 images we have.
+Save the layer (using shapefile format) using EPSG:32633 as the CRS, to match the same CRS as the Sentinel-2 images we have.
 === Clipping polygon ===