06_convolutions.ipynb

Kernel: Python 3

View source on GitHub

Notebook Viewer

Run in Google Colab

Convolutions

To perform linear convolutions on images, use image.convolve(). The only argument to convolve is an ee.Kernel which is specified by a shape and the weights in the kernel. Each pixel of the image output by convolve() is the linear combination of the kernel values and the input image pixels covered by the kernel. The kernels are applied to each band individually. For example, you might want to use a low-pass (smoothing) kernel to remove high-frequency information. The following illustrates a 15x15 low-pass kernel applied to a Landsat 8 image:

Install Earth Engine API and geemap

Install the Earth Engine Python API and geemap. The geemap Python package is built upon the ipyleaflet and folium packages and implements several methods for interacting with Earth Engine data layers, such as Map.addLayer(), Map.setCenter(), and Map.centerObject(). The following script checks if the geemap package has been installed. If not, it will install geemap, which automatically installs its dependencies, including earthengine-api, folium, and ipyleaflet.

Important note: A key difference between folium and ipyleaflet is that ipyleaflet is built upon ipywidgets and allows bidirectional communication between the front-end and the backend enabling the use of the map to capture user input, while folium is meant for displaying static data only (source). Note that Google Colab currently does not support ipyleaflet (source). Therefore, if you are using geemap with Google Colab, you should use import geemap.foliumap. If you are using geemap with binder or a local Jupyter notebook server, you can use import geemap, which provides more functionalities for capturing user input (e.g., mouse-clicking and moving).

# Installs geemap package
import subprocess

try:
    import geemap
except ImportError:
    print("geemap package not installed. Installing ...")
    subprocess.check_call(["python", "-m", "pip", "install", "geemap"])

# Checks whether this notebook is running on Google Colab
try:
    import google.colab
    import geemap.foliumap as emap
except:
    import geemap as emap

# Authenticates and initializes Earth Engine
import ee

try:
    ee.Initialize()
except Exception as e:
    ee.Authenticate()
    ee.Initialize()

Create an interactive map

The default basemap is Google Satellite. Additional basemaps can be added using the Map.add_basemap() function.

Map = geemap.Map(center=[40, -100], zoom=4)
Map.add_basemap("ROADMAP")  # Add Google Map
Map

Add Earth Engine Python script

# Load and display an image.
image = ee.Image("LANDSAT/LC08/C01/T1_TOA/LC08_044034_20140318")
Map.setCenter(-121.9785, 37.8694, 11)
Map.addLayer(image, {"bands": ["B5", "B4", "B3"], "max": 0.5}, "input image")

# Define a boxcar or low-pass kernel.
# boxcar = ee.Kernel.square({
#   'radius': 7, 'units': 'pixels', 'normalize': True
# })

boxcar = ee.Kernel.square(7, "pixels", True)

# Smooth the image by convolving with the boxcar kernel.
smooth = image.convolve(boxcar)
Map.addLayer(smooth, {"bands": ["B5", "B4", "B3"], "max": 0.5}, "smoothed")

Map.addLayerControl()
Map

The output of convolution with the low-pass filter should look something like Figure 1. Observe that the arguments to the kernel determine its size and coefficients. Specifically, with the units parameter set to pixels, the radius parameter specifies the number of pixels from the center that the kernel will cover. If normalize is set to true, the kernel coefficients will sum to one. If the magnitude parameter is set, the kernel coefficients will be multiplied by the magnitude (if normalize is also true, the coefficients will sum to magnitude). If there is a negative value in any of the kernel coefficients, setting normalize to true will make the coefficients sum to zero.

Use other kernels to achieve the desired image processing effect. This example uses a Laplacian kernel for isotropic edge detection:

Map = geemap.Map(center=[40, -100], zoom=4)

# Define a Laplacian, or edge-detection kernel.
laplacian = ee.Kernel.laplacian8(1, False)

# Apply the edge-detection kernel.
edgy = image.convolve(laplacian)
Map.addLayer(edgy, {"bands": ["B5", "B4", "B3"], "max": 0.5}, "edges")
Map.setCenter(-121.9785, 37.8694, 11)
Map.addLayerControl()
Map

Note the format specifier in the visualization parameters. Earth Engine sends display tiles to the Code Editor in JPEG format for efficiency, however edge tiles are sent in PNG format to handle transparency of pixels outside the image boundary. When a visual discontinuity results, setting the format to PNG results in a consistent display. The result of convolving with the Laplacian edge detection kernel should look something like Figure 2.

There are also anisotropic edge detection kernels (e.g. Sobel, Prewitt, Roberts), the direction of which can be changed with kernel.rotate(). Other low pass kernels include a Gaussian kernel and kernels of various shape with uniform weights. To create kernels with arbitrarily defined weights and shape, use ee.Kernel.fixed(). For example, this code creates a 9x9 kernel of 1’s with a zero in the middle:

# Create a list of weights for a 9x9 kernel.
list = [1, 1, 1, 1, 1, 1, 1, 1, 1]
# The center of the kernel is zero.
centerList = [1, 1, 1, 1, 0, 1, 1, 1, 1]
# Assemble a list of lists: the 9x9 kernel weights as a 2-D matrix.
lists = [list, list, list, list, centerList, list, list, list, list]
# Create the kernel from the weights.
kernel = ee.Kernel.fixed(9, 9, lists, -4, -4, False)
print(kernel.getInfo())