Blending visualizations#

You can blend manually or automatically.

Manual blending allows you to use visualizations that are not part of rvt. When blending manually you have to define each layer (visualization) in Python.

Automatic blending automatically computes rvt visualizations and blends them together according to a configuration JSON file, which can be edited.

The main class of the rvt.blend module for blending is BlenderCombination, which has the list attribute layers where instances of class BlenderLayer are stored. In BlenderLayer instances in layers we store a specific visualization and its parameters for blending.

The BlenderCombination class has the method render_all_images(), which blends together all BlenderLayer instances (visualizations) in the BlenderCombination.layers list and outputs the blended image.

You can blend as many layers as you want.

Manual blending#

Example

Let’s say we have already calculated the simple local relief model (slrm_arr), slope (slope_arr) and hillshade (hillshade_arr), and now need to blend all the calculated visualizations together:

# import the module
import rvt.blend

# create the BlenderCombination() class instance which will hold the layers (visualizations)
combination_manual = rvt.blend.BlenderCombination()

# call BlenderCombination.create_layer() to add a layer
# this creates a BlenderLayer instance and adds it to BlenderCombination.layers

# 1st layer
# add slrm layer with 2% perc cuttoff on both sides, multiply blend mode and 25% opacity
combination_manual.create_layer(
    vis_method="Simple local relief model",
    normalization="perc",
    minimum=2,
    maximum=2,
    blend_mode="multiply",
    opacity=25,
    image=slrm_arr
    )

# 2nd layer
# add slope layer with value stretch from 0 to 51, luminosity blend mode and 50% opacity
combination_manual.create_layer(
    vis_method="Slope gradient",
    normalization="value",
    minimum=0,
    maximum=51,
    blend_mode="luminosity",
    opacity=50,
    image=slope_arr
    )

# 3rd layer
# add hillshade layer with value stretch from 0 to 1, normal blend mode and 100% opacity
combination_manual.create_layer(
    vis_method="Hillshade",
    normalization="value",
    minimum=0,
    maximum=1,
    blend_mode="normal",
    opacity=100,
    image=hillshade_arr
    )

# if we want to save the blended image to a file, we need to add dem_path to the
# combination (for metadata, geodata)
combination_manual.add_dem_path(dem_path=input_dem_path)

# blend them all together
# you can save the blend to GeoTIFF if save_render_path presented
# (and dem_path is added), otherwise it only returns array
render_arr = combination_manual.render_all_images(save_render_path=output_blend_path)

Example

You can also let the BlenderCombination class automatically compute the visualization or give the path to a visualization.

If you don’t provide the image parameter, and the vis_method parameter is correct (an existing rvt.vis function), blender automatically calculates the visualization.

If you don’t provide the image parameter, but do provide the image_path parameter (if you provide both image will be used), blender will read the visualization from image_path.

If you don’t provide the image and image_path parameters, you have to add an rvt.default.DefaultValues instance as a parameter to BlenderCombination.render_all_images(). Blender then takes the parameters set in this class when calculating specific visualizations. You also have to add dem array and its resolution.

The example below uses all three methods:

# import all required modules
import rvt.blend
import rvt.default

# create the BlenderCombination() class instance which will hold the layers (visualizations)
combination_manual = rvt.blend.BlenderCombination()

# we will let blender compute the slrm visualization. so, we need to create
# rvt.default.DefaultValues() and change the parameters for slrm. we will later
# add default to the combination_manual.render_all_images() method
default = rvt.default.DefaultValues()
default.slrm_rad_cell = 15

# 1st layer
# add slrm layer with 2% perc cuttoff on both sides, multiply blend mode and 25% opacity
# image and image_path parameters both not provided, so slrm is calculated automatically
combination_manual.create_layer(
    vis_method="Simple local relief model",
    normalization="perc",
    minimum=2,
    maximum=2,
    blend_mode="multiply",
    opacity=25
    )

# 2nd layer
# add slope layer with value stretch from 0 to 51, luminosity blend mode and 50% opacity
# image_path parameter provided to slope, so slope is read from file
combination_manual.create_layer(
    vis_method="Slope gradient",
    normalization="value",
    minimum=0,
    maximum=51,
    blend_mode="luminosity",
    opacity=50,
    image_path=slope_path
    )

# 3rd layer
# add hillshade layer with value stretch from 0 to 1, normal blend mode and 100% opacity
# image parameter provided
combination_manual.create_layer(
    vis_method="Hillshade",
    normalization="value",
    minimum=0,
    maximum=1,
    blend_mode="normal",
    opacity=100,
    image=hillshade_arr
    )

# we have to add dem array and resolution so that slrm can be computed
combination_manual.add_dem_arr(dem_arr=input_dem_arr, dem_resolution=resolution)

# blend them all together and add default where slrm parameters are defined
render_arr = combination_manual.render_all_images(default=default)

Automatic blending#

Automatic blending is blending from a configuration JSON file. You can create a JSON file and change it to suit your needs.

Example

# import the module
import rvt.blend

# create the BlenderCombination() class
combination_auto = rvt.blend.BlenderCombination()

# to create the JSON blender combination configuration file example, change the
# path to where you wish to save the file
blender_combination_path = r"settings\blender_file_example.txt"
rvt.blend.create_blender_file_example(file_path=blender_combination_path)

# set the parameters of the visualizations you will be using
default = rvt.default.DefaultValues()
# for example default.hs_sun_el=40

# read the JSON combination configuration file
combination_auto.read_from_file(file_path=blender_combination_path)

# needed when save_visualizations is True and save_render_path is not None
layers_auto.add_dem_path(input_dem_path)

# call the method render_all_images() and its parameters
# we can save a specific visualization (to dem_path directory) if we set the
# parameter ``save_visualization`` to True
layers_auto.render_all_images(
    default=default,
    save_visualizations=True,
    save_render_path=output_blend_path,
    save_float=True,
    save_8bit=True # set save_8bit=True if you also wish to save an 8bit version
    )