3D Point Cloud Model of COVID-19

This project will create a 3D point cloud visualization model of the COVID-19 pandemic. The model will focus on modeling the global spatial patterns along with the statistics of different types of live cases for each country by thousands or even millions of georeferenced points.

Preparation

Check python module/package installation. Uncomment the pip statement to install first if any error occurs through import step.

STEP1: LAND MASK DATA

You do not need to run step 1 again because the local landmask.geojson has already created. But you are welcomed to uncomment these code cells to run again if needed.

1. Use numpy to populate 2 1-D arrays for Longitude and Latitude respectively. Then use Cartesian Product to create a 2D array representing the Longitude and Latitude pairs of grid points.

2. Convert the 2D array to a temporary dictionary and then load it to Pandas Dataframe.

3. Convert the Pandas Dataframe to GeoPandas Dataframe by creating geometry from the lon and lat values.

4. Load the shapefile of World Boundaries and create a GeoPandas Dataframe.

5. Perform a spatial join between the grid points and the world polygons, and make each point have the values for Country Name, ISO and location information.

6. Write the spatial join result to a GeoJSON file and save locally.

STEP2: COVID-19 DATA

You do not need to run step 2 again because the local iso_coords.json has already created. But you are welcomed to uncomment these code cells to run again if needed.

1. Read the local landmask.geojson file, and extract the ISO and corresponding geometry. (There are 172 countries covered in the file.)

2. Request the available countries that have been kept tracking from the COVID-19 API. (There are 248 countries kept tracking on the API.)

3. COVID-19 data are stored individually by country with different slug mentioned in the request url. Thus, extract the slug for the 172 countries and store in iso_slugs variable.

4. Loop iso_slugs dict and request each country url. Modify the response data by removing usless keys and adding other keys. Store the modified data in local folder data_by_country individually. Also provide the request date on the filename.

STEP3: MODEL DEVELOPMENT

1. Read local COVID-19 data and store in covid_coords list. The list involves the coordinates(lon/lat) and case(e.g. latest active cases) for each georeforenced point.

2. Convert the coordinates from Geodetic WGS84(lon/lat) to 3-dimensional Geocentric Cartesian Coordinates(x/y/z). Points are set on the surface of the globe.

3. Use colormap to visualize the latest active cases data. In order to have more obvious color distributions, find the minimum and maximum observation values after leaving out the dirty data.

4. Construct the formatted 3D Polygon File Format(PLY) with coordinates and colors for each georeferenced point.