User Guide

Quick Start

  1. Click icon in the QGIS Tool Bar or via Raster ‣ EO Time Series Viewer to start the EO Time Series Viewer

  2. Click Files ‣ Add example to load an exemplary time series of Landsat and RapidEye observations.

    Open the sensor panel View ‣ Panels… ‣ Sensors / Products and change the sensor names

    Default Changed
    6bands@30.0m Landsat
    5bands@5.0m RapidEye3

    The change of sensor names becomes visible in other other parts of the GUI as well, e.g. the list of time series source images.

  1. Use the scroll band and the map tools mActionZoomIn mActionZoomOut mActionPan to change the temporal and spatial subset shown from the time series.
  2. Open the Mapping panel, change the map view name from Map View 1 to True Color and set the multiband color render band selection to RGB = 3-2-1 for both, Landsat and RapidEye images (right-click on layer, Set Properties ‣ Style).

Note

Use mActionRefresh to refresh or redraw related maps, spectral profile plots etc.

  1. Now we like to optimize the color stretch. Choose a -clouded Landsat observation like 2014-06-24 and use the map context menu (right-mouse-click) to click on Stretch to current extent … ‣ Linear 5%. Repeat with Linear and Gaussian stretch as well as for RapidEye images to see how this changes the band-specific min/max values in the Map View settings.

  2. Click mActionAddMapView to create a second map view, respectively row of map images. Call it Short-Wave IR and the the multiband color render bands to Landsat RGB = 4-5-3 and RapidEye RGB = 5-4-3

  3. Expand the Map Properties combobox (in the Mapping panel), increase the map size to 300x300 pixel and press Apply.

  4. Bring the Temporal Profile panel View ‣ Panels… ‣ Sensors / Products and the 2D page to the front. Click mIconTemporalProfile and select a map location to visualize the temporal profile for. Each selected map location will be listed in the panels mActionOpenTable Coordinate page.

    Go back to the 2D plot page and add a second profile with RapidEye data as Sensor source. Change the expression field to show the Normalized Difference Vegetation Index (NDVI) for both sensors:

    Sensor Expression NDVI
    Landsat (b4-b3)/(b4+b3)
    RapidEye (b5-b3)/(b5+b3)


The Graphical User Interface

_images/interface.png

This is what the EO Time Series Viewer’s interface looks like when opening it.

Note

Just like in QGIS, many parts of the GUI are adjustable panels. You can arrange them as tabbed, stacked or separate windows. Activate/Deactivate panels under View ‣ Panels


Time Series

This window lists the individual input raster files of the time series.

_images/timeseriespanel.png
  • Date corresponds to the image acquisition date as automatically derived by the EO TSV from the file name. Checking cbc or unchecking cbu the box in the date field will include or exclude the respective image from the display
  • Sensor shows the name of the sensor as defined in the Sensors / Products tab
  • ns: number of samples (pixels in x direction)
  • nl: number of lines (pixels in y direction)
  • nb: number of bands
  • image: path to the raster file

You can add new rasters to the time series by clicking mActionAddRasterLayer Add image to time series. Remove them by selecting the desired rows in the table (click on the row number) and pressing the mActionRemoveTSD Remove image from time series button.

Tip

If you have your time series available as one large raster stack, you can import this file via Files ‣ Add images from time stack

Tip

Click Files ‣ Add example to load a small example time series.


Sensors / Products

_images/sensordock.png

The EO Time Series Viewer automatically assesses different characteristics of the input images (number of bands, geometric resolution etc.) and combines identical ones into sensor groups (or products). Those are listed as follows in the Sensor / Products window:

  • name is automatically generated from the resolution and number of bands (e.g. 6bands@30.m). This field is adjustable, i.e. you can change the name by double-clicking into the field. The here defined name will be also displayed in the Map View and the Time Series table.
  • nb: number of bands
  • n images: number of images within the time series attributed to the according sensor
  • wl: comma separated string of the (center) wavelength of every band and [unit]
  • id: string identifying number of bands, geometric resolution and wavelengths (primary for internal use)

The Toolbar

Button Function
mActionAddRasterLayer Add images to the time series
mActionAddTS Add Time Series from CSV
mActionRemoveTS Remove all images from Time Series
mActionSaveTS Save Time Series as CSV file
mActionAddOgrLayer Add vector data file
qgsMapCenter Synchronize with QGIS map canvas
mActionAddMapView Add maps that show a specified band selection
mActionRefresh Refresh maps
mActionPan Pan map
mActionZoomIn Zoom into map
mActionZoomOut Zoom out
mActionZoomActual Zoom to pixel scale
mActionZoomFullExtent Zoom to maximum extent of time series
pan_center Center map on clicked locations
select_location Identify Pixels and Features
mActionPropertiesWidget Identify cursor location values
profile Identify raster profiles to be shown in a Spectral Library
mIconTemporalProfile Identify pixel time series for specific coordinate
mActionSelectRectangle Select Features
mActionToggleEditing Start Editing Mode
mActionSaveEdits Save Edits
mActionCapturePolygon Draw a new Feature

Note

Only after select_location Identify Pixels and Features is activated you can select the other identify tools (mActionPropertiesWidget, profile, mIconTemporalProfile). You can activate them all at once as well as of them, in case of the latter variant clicking in the map has no direct effect (other than moving the crosshair, when activated)


Map Visualization

Map Properties

In the map properties box you can specify Width and Height, as well as background Color and the CRS of the single map canvases. Click Apply to apply changes. By default the keep ratio option is cbc checked, i.e. height will be the same as width. In case you want to have unequally sized views, deactivate this option.

_images/mapproperties.png

Map Views

A map view is a row of map canvases that show the time series images of different sensors/product in the same band combination, e.g. as “True Color bands”. The map view panel allows to add or remove map views and to specifiy how the images of each sensor are to be rendered.

  • You can add new Map Views using the mActionAddMapView button. This will create a new row of map canvases. Remove a map view with the mActionRemoveMapView button.
  • In case the Map View does not refresh correctly, you can ‘force’ the refresh using the mActionRefresh button (which will also apply all the render settings).
  • Access the settings for individual Map Views by clicking in the mapview mapviewbutton
  • You can use the questionmark button to highlight the current Map View selected in the dropdown menu (respective image chips will show red margin for a few seconds).

For every Map View you can alter the following settings:

  • Hide/Unhide the Map View via the mapviewHidden Toggle visibility of this map view button.
  • Activate/Deactivate Crosshair via the crosshair Show/hide a crosshair button. Press the arrow button next to it to enter the Crosshair specifications symbology , where you can customize e.g. color, opacity, thickness, size and further options.
  • You may rename the Map View by altering the text in the Name field.

Layer representation:

  • Similar to QGIS you can change the visual representation of raster or vector layers in the layer properties. To open them, right-click on the layer you want to alter and select Set Properties ‣ Style

    _images/layerproperties.png

Cursor Location Values

This tools lets you inspect the values of a layer or multiple layers at the location where you click in the map view. To select a location (e.g. pixel or feature) use the select_location Select Cursor Location button and click somewhere in the map view.

  • The Cursor Location Value panel should open automatically and list the information for a selected location. The layers will be listed in the order they appear in the Map View. In case you do not see the panel, you can open it via View ‣ Panels ‣ Cursor Location Values.

    _images/cursorlocationvalues.png

  • By default, raster layer information will only be shown for the bands which are mapped to RGB. If you want to view all bands, change the Visible setting to All (right dropdown menu). Also, the first information is always the pixel coordinate (column, row).

  • You can select whether location information should be gathered for All layers or only the Top layer. You can further define whether you want to consider Raster and Vector layers, or Vector only and Raster only, respectively.

  • Coordinates of the selected location are shown in the x and y fields. You may change the coordinate system of the displayed coordinates via the mActionSetProjection Select CRS button (e.g. for switching to lat/long coordinates).


Profile Visualization

_images/example_2dprofile.png

Example: Temporal NDSI (Normalized Difference Snow Index) profile for 2 locations using Landsat 7 and 8 images.

Temporal Profiles

The Temporal Profiles panel lets you visualize temporal profiles. On the left side you can switch between the mIconTemporalProfile2D profile and the coordinates mActionOpenTable page. The latter lists all coordinates of selected or imported profile locations.

Adding and managing a temporal profile:

  • You can use the mIconTemporalProfile button to click on a location on the map an retrieve the temporal profile, or in the toolbar select select_location + mIconTemporalProfile.

  • Mind how the selected pixel now also appears on the coordinates mActionOpenTable page!

  • If you select further pixels ( mIconTemporalProfile ), they will be listed in the coordinates page, but not automatically visualized in the plot.

  • Use mActionAdd to create an additional plot layer, and double-click in the Coordinate field in order to select the desired location (so e.g. the newly chosen pixel) or just change the location in the current plot layer.

  • Similarly, you can change the sensor to be visualized by double-clicking inside the Sensor field and choosing from the dropdown.

  • Click inside the Style field to change the visual representation of your time series in the plot.

  • Remove a time series profile by selecting the desired row(s) and click mActionRemove.

  • The DN or Index field depicts which values will be plotted.

    • Here you may select single bands (e.g. b1 for the first band)

    • or you can calculate indices on-the-fly: e.g. for the Landsat images in the example dataset the expression (b4-b3)/(b4+b3) would return the NDVI.

      _images/example_temppindex.png

      Example of visualizing the NDVI for the same location for different sensors (example dataset).

  • You can also move the map views to a desired date from the plot directly by right-click into plot ‣ Move maps to …

Note

The EO TSV won’t extract and load all pixel values into memory by default in order to reduce processing time (only the ones required). You can manually load all the values by clicking the mIconTemporalProfileRefresh Load missing band values button on the coordinates mActionOpenTable page.

Importing or exporting locations:
  • You can also import locations from a vector file instead of collecting them from the map: Go to the coordinates mActionOpenTable page and add locations via the mActionAddOgrLayer button.
  • If you want to save your locations, e.g. as shapefile or CSV, click on mActionFileSave.

Spectral Library

The spectral library view allows you to visualize, label and export spectral profiles.

_images/spectrallibrarypanel.png
  • Use the pickrasterspectrum Select a spectrum from a map button to extract and visualize a pixels profile (by clicking on a pixel on the map).
  • You can add a selected spectrum to your spectral library by clicking on profile2speclib.
  • The gathered spectra are listed in the table on the right. For every spectrum additional metadata will be stored, e.g. the date, day of year and sensor.
  • When the profile2speclib_auto button is activated, the profile will be directly added to the library after clicking on a pixel.
  • Change the display style (color, shape, linetype) in the Spectral Library Properties, which can be accessed via the action button in the lower right.

Note

The spectral library table behaves quite similar to the attribute table you know from QGIS:
  • You can edit the content by entering the editing mode mActionToggleEditing
  • You can add further information by adding fields via the mActionNewAttribute button (e.g. different class labels). Remove them with mActionDeleteAttribute, accordingly.
  • Double-click into a desired field to change its content
  • Remove spectra by selecting the desired row(s) in the table and click mActionDeleteSelected

Quick Labeling

The EO Time Series Viewer assists you in describing, i.e. label, reference data, e.g. to describe at the occurence of land cover types and events. Whether your locations (point, lines or polygons) of interest already exist, or are being digitized in the visualized maps first, in both cases you want to describe them in the attribute table of the vector layer.

The EO Time Series Viewer supports this with “Quick Label” short-cuts from the map context menu. If triggered, they send temporal or categorical information direct to the attribute cells of selected geometries, e.g a polygon that has been selected in a map.

_images/quicklabel_workflow.png

Let’s assume we have a map related to a Landsat (LND) observation taken at 2019-02-05T11:23:42.00. The following table shows how this is translated into linked vector fields:

Mapping of Quick Labels information into vector layer attributes
  Type of linked vector layer field
LabelType varchar int double datetime date time bool blob
Date 2019-02-05     2019-02-05T00:00:00 2019-02-05      
Date-Time 2019-02-05T11:23:42     2019-02-05T11:23:42 2019-02-05      
Time 11:23:42     2019-02-05T11:23:42   11:23:42    
Day of Year (DOY) 36 36 36          
Year 2019 2019   2019-02-05T11:23:42 2019-02-05 11:23:42    
Decimal Year 2019.0980926430518 2019 2019.098          
Sensor Name LND              
Source Image /path/to/image              

To use Quick Labels, open the vector layer properties and activate them in the Attribute Form widget:

_images/quicklabel_settings.png