netCDF polygons conversion to fixed-sized raster image

Question

I'm trying to convert data from netCDF files; the data contains polygons and values for methane concentrations.

from matplotlib.patches import Polygon
import cartopy.crs as ccrs
import cartopy.feature as cfeature
# %% Plotting the polygons
fig, ax = plt.subplots(figsize=(10, 10), subplot_kw={'projection': ccrs.PlateCarree()})
ax.set_extent([-9, -5, 32, 34], crs=ccrs.PlateCarree())
# Add a base map
ax.add_feature(cfeature.LAND)
ax.add_feature(cfeature.OCEAN)
ax.add_feature(cfeature.COASTLINE)
ax.add_feature(cfeature.BORDERS, linestyle=':')
# Normalize the ch4 values for the colormap
norm = mcolors.Normalize(vmin=1850, vmax=1880)
cmap = matplotlib.colormaps['rainbow']
# Plot the grid cells and color them based on ch4 values
for lat_corners, lon_corners, ch4 in zip(lat_corners, lon_corners, ch4_values):
 color = cmap(norm(ch4))
 poly = Polygon(list(zip(lon_corners, lat_corners)), facecolor=color,
 edgecolor='grey', linewidth=0.5, alpha=0.7)
 ax.add_patch(poly)
# Add a colorbar
sm = plt.cm.ScalarMappable(cmap=cmap, norm=norm)
sm.set_array([])
cbar = plt.colorbar(sm, ax=ax, orientation='horizontal', fraction=0.046, pad=0.04)
cbar.set_label('Methane Concentration (ppb)')
plt.show()

polygons with methane concentration

This image is very specific, and I want to apply this to generate a dataset with the shape of [N,M,32x32] pixel images, where N is the number of images and M contains different forms of data (methane values, wind, etc.).

One of the big issues I've been having is treating spaces between polygons (where one is not there) as a pixel.

I apologize for any lack of information; this is something very new to me.

def project_to_image(lat_corners, lon_corners, ch4_values):
 # Create a 32x32 empty image
 #image = np.zeros((32, 32, 3), dtype=np.uint8)
 #image = np.full((32, 32, 3), 255, dtype=np.uint8)
 norm = mcolors.Normalize(vmin=1850, vmax=1880)
 cmap = matplotlib.colormaps['rainbow']
 #min_value_color = cmap(norm(1850)) # Color for the minimum value in the range
 #min_value_color = (np.array(min_value_color[:3]) * 255).astype(np.uint8) # Convert to RGB
 min_value_color = 255
 # Create a 32x32 empty image with the background color set to the minimum value color
 image = np.full((32, 32, 3), min_value_color, dtype=np.uint8)
 # Create a Basemap instance
 m = Basemap(projection='cyl', llcrnrlat=31.5, urcrnrlat=34.0, llcrnrlon=-9.0, urcrnrlon=-5.0, resolution='h')
 for lat_corner, lon_corner, ch4_value in zip(lat_corners, lon_corners, ch4_values):
 if np.any(~lon_corner.mask) and np.any(~lat_corner.mask):
 color = cmap(norm(ch4_value))
 x, y = m(lon_corner[~lon_corner.mask], lat_corner[~lat_corner.mask])
 # Limit coordinates to Basemap range
 x = np.clip(x, m.llcrnrx, m.urcrnrx)
 y = np.clip(y, m.llcrnry, m.urcrnry)
 # Adjust coordinates to fit 32x32 image
 # Map coordinates to 0-31 range
 x = (31*(x - m.llcrnrx) / (m.urcrnrx - m.llcrnrx) ).astype(int)
 y = (31*(y - m.llcrnry) / (m.urcrnry - m.llcrnry) ).astype(int)
 # Filter out of range coordinates
 #valid_idx = (x >= 0) & (x < 32) & (y >= 0) & (y < 32)
 #x = x[valid_idx]
 #y = y[valid_idx]
 # Draw polygon
 for i in range(len(x)):
 image[int(y[i]), int(x[i])] = (np.array(color[:3]) * 255).astype(np.uint8)
 return image

output for the above code

what i am trying to do:

This is from the research article.

The research article that I am trying to replicate the data from can be found here:

Research article

• gboffi: I cannot understand your problem, and of course I'm thick but maybe you can edit your post to clarify the question. By the way, I cannot see a single question mark in your post :-)
• sapphire: I know, a lot of edits were done by others on my wording of the problem that didn't help in my opinion, just made it vague. I fortunately recently solved this issue and have been preparing a way to answer it on here. My main issue was taking the values in the dataset, such as the polygons which vary in size and projecting them to a grid where each variable (methane concentration, etc.) is a separate matrix to a tensor (which represents one scene). I did a write up on it medium.com/@aamavizca/…