Skip to main content

EarthAI Platform Support Portal

Visualizing Assets with %%eod Cell Magic

Courtney Whalen
  • Updated

In this tutorial, we demonstrate how to use %%eod cell magic for different viewing options of assets returned by an EarthAI Catalog query.

We show four different viewing options available, which include:

  1. --extents, which displays scene footprints on a map;
  2. --heatmap, which renders a heatmap of scene centroids;
  3. --hist, which creates a temporal density histogram of results; and
  4. --table, which shows a table of results with tile samples.

We also demonstrate the --side option, which places map widgets in a sidebar to the right of the notebook rather than in a cell output. To see the full list of parameter options, run ?eod in a notebook cell.

We use Landsat 8 imagery covering Charlottesville, Virginia, USA in each query.

View Scene Footprints

By adding the --extents option, you can view the scene boundaries of the assets returned by the %%eod query on an ipyleaflet map.

You can hover the mouse pointer over any one of the map extents to display a thumbnail of the image and view the asset information.

In [1]:
%%eod --extents
{
    "collections": [
        "landsat8_l1tp"
    ],
    "geo": [
        -78.57147216796875,
        37.96152331396614,
        -78.40942382812501,
        38.09133660751176
    ],
    "max_cloud_cover": 100
}
cell1.png

View a Heatmap of Scenes

By adding the --heatmap option, you can view a heatmap of scene centroids. Lighter colors (yellow) indicates more scene coverage and darker colors (purple) indicate less scene coverage. Additionally, you can add an integer argument to --heatmap, which sets the kernel radius (in pixels) of the heatmap.

In [2]:
%%eod --heatmap 100
{
    "collections": [
        "landsat8_l1tp"
    ],
    "geo": [
        -78.57147216796875,
        37.96152331396614,
        -78.40942382812501,
        38.09133660751176
    ],
    "max_cloud_cover": 100
}
cell2.png

View a Temporal Density Histogram

By adding the --hist option, you can display a temporal density histogram of your %%eod query results. This will return the number of scenes at each datetime interval that meet your query parameters.

In the upper right corner of the histogram, there's a button that allows you to save the histogram as a PNG or SVG.

In [3]:
%%eod --hist
{
    "collections": [
        "landsat8_l1tp"
    ],
    "geo": [
        -78.57147216796875,
        37.96152331396614,
        -78.40942382812501,
        38.09133660751176
    ],
    "max_cloud_cover": 100
}
cell3.png

View Results in a Table

By adding the --table option, you can view a sample of results returned from the %%eod query. Additionally, you can specify the number of rows to display if you pass an integer argument along with the --table option as shown below.

The table will include a natural color preview of each scene as well as a small map of scene boundaries highlighting where the scene is located over a basemap.

In [4]:
%%eod --table 2
{
    "collections": [
        "landsat8_l1tp"
    ],
    "geo": [
        -78.57147216796875,
        37.96152331396614,
        -78.40942382812501,
        38.09133660751176
    ],
    "max_cloud_cover": 100
}
cell4.png

View Map Widgets in a Side Panel

By specifying --side as an option in %%eod magic, the map widgets will show up in a right sidebar next to the notebook rather than in the cell output.

As shown below, the extents and heatmap will show up on a single map widget in the sidebar while the histogram and table will show up in the cell output.

In [ 5]:
%%eod --extents --heatmap 100 --hist --table 2 --side
{
    "collections": [
        "landsat8_l1tp"
    ],
    "geo": [
        -78.57147216796875,
        37.96152331396614,
        -78.40942382812501,
        38.09133660751176
    ],
    "max_cloud_cover": 100
}

Related articles

Comments

0 comments

Please sign in to leave a comment.