Customising Your Map Layer Visualisation: How to Guide

This article explains how to use the customisation feature of the Add Map Layer block.

TABLE OF CONTENTS

• Make sure you understand the Add Map Layer block
• Customising when using multiple band visualisation (e.g. an RGB composite)
• Customising when using a single-band visualisation
• How Earth Blox determines the contrast settings
  • Contrast settings for image collections (time series)
  • Contrast settings for a single image

Make sure you understand the Add Map Layer block

• To visualise data on the map, you need to include an Add Map Layer block.
• It is a dynamic block, so modifies its appearance depending on the other blocks within the workflow. 
• For example, when you add it to a dataset container block with Sentinel 2 selected, the choices look like this:

• You can choose to view only a single band or an RGB colour composite. 
• For each of the three colour bands (RGB) the drop-down menu gives you a choice of which bands are available.
• Once you have selected the bands, you can customise the brightness, contrast, scaling, etc, by clicking on the settings icon (the cog-wheel).

Customising when using multiple band visualisation (e.g. an RGB composite)

• When you choose to view multiple bands, the colour of the image is defined by the proportional values of the RGB bands so choosing a colour palette is not relevant.  
• You will be offered the choice to adjust the Contrast Options.
• The Contrast Options are for defining the minimum and maximum value to scale your images. You are given the following choices:
  • Default Dataset Visualisation:  The minimum and maximum scaling is pre-defined as a generally good setting for the data selected.
  • Dataset Min and Max: The min and max scaling is set to the lowest and highest values within the dataset.
  • Mean +/- 1 Standard Deviation:  The min and max are defined by the mean of the data plus and minus the standard deviation of the data. (If the data are normally distributed this will include approximately 68% of all data).
  • Mean +/- 2 Standard Deviations: The min and max are defined by the mean of the data plus and minus two standard deviations of the data. (If the data are normally distributed, this will include approximately 95% of all data).
  • p90: Scales the data so that min=0 but with max set so that the scaling includes 90% of all the data. (This is used to ensure that infrequent but very high values don't bias the scaling). 
  • p98: Scales the data so that min=0 but with max set so that the scaling includes 98% of all the data. (This is used to ensure that very infrequent but extremely high values don't bias the scaling). 
  • Custom: This allows you to specify your own minimum and maximum value for each band.  This means that for each of the Red, Green and Blue channels, you can apply different scaling. 
• Note that when analysing a time series, the contrast statistics are calculated for the middle image (by time) and then applied to all of the others. 
• The transparency of a layer on the map is governed by a slider on the Areas of Interest box on the map window, not on the Add Map Layer block.

Customising when using a single-band visualisation

• When you choose to view only a single band, when you click on the settings icon you will see two options:
  • Choose Colour Ramp.
  • Contrast Options.
• The Colour Ramp is a selection of pre-defined colour palettes.  Click on the drop-down menu to see the selection.
• The Invert toggle switch reverses the direction of the palette so that the max and min are the other way around. 
• The Contrast Options are for defining the minimum and maximum value to scale your images. You are given the following choices:
  • Default Dataset Visualisation:  The minimum and maximum scaling is pre-defined as a generally good setting for the data selected.
  • Dataset Min and Max: The min and max scaling is set to the lowest and highest values within the dataset.
  • Mean +/- 1 Standard Deviation:  The min and max are defined by the mean of the data plus and minus the standard deviation of the data. (If the data are normally distributed this will include approximately 68% of all data).
  • Mean +/- 2 Standard Deviations: The min and max are defined by the mean of the data plus and minus two standard deviations of the data. (If the data are normally distributed, this will include approximately 95% of all data).
  • p90: Scales the data so that min=0 but with max set so that the scaling includes 90% of all the data. (This is used to ensure that infrequent but very high values don't bias the scaling). 
  • p98: Scales the data so that min=0 but with max set so that the scaling includes 98% of all the data. (This is used to ensure that very infrequent but extremely high values don't bias the scaling). 
  • Custom: This allows you to specify your own minimum and maximum value.  
• Note that when analysing a time series, the contrast statistics are calculated for the middle image (by time) and then applied to all of the others. 
• The transparency of a layer on the map is governed by a slider on the Areas of Interest box on the map window, not on the Add Map Layer block.

How Earth Blox determines the contrast settings

Automatically determining contrast values can be very computationally expensive, and therefore slow. Earth Blox uses two methods for doing this, designed to provide a high level of accuracy, whilst not significantly slowing the workflow run.

Contrast settings for image collections (time series)

Contrast settings for a collection of images in Earth Blox are calculated via a random sample of 1000 pixels per image.

This has two potential consequences:

• If a workflow is re-run, a different sample would be taken, and the contrast values may change slightly
• The contrast values are not guaranteed to be the universal values for an image as they are based on sampling. So an image may have pixel values ranging from -1.00 to +1.00, but selecting the "Dataset Min and Max" option might scale the data as -0.98 to +0.99. If you know the min and max pixel values (such as with an NDVI), then these can be set via the "Custom" option.

Contrast settings for a single image

For a single image, a different approach is taken to calculate contrast values. Earth Blox will use all pixels in the image, at the highest resolution possible that keeps the total number of pixels below 10,000.

We always track the feedback you give us so that we can improve how we help you. 

Please let us know if you found this one helpful.