Visualizing Collection of Pixels as Image in Google Earth Engine

Question

I have a polygon and I want to extract both the value and coordinates of every pixel of a landcover raster within that polygon into a CSV spreadsheet. I will use that to perform computations outside GEE that result in a label assigned to each pixel (an extra column in the CSV). What I want to do is to both plot an image of the pixel values outside a map (e.g., using Matplotlib) and to view the image on GEE. For this, I use sampleRegions to give me a featureCollection of all pixels within the polygon, export that, and add the new label column to it. However, I don’t know how to visualize the resulting updated featureCollection as an image. How can I know where each pixel goes in the image?

Here is my code (I assign a random label to each pixel for demonstration):

import ee
ee.Initialize()
from random import random

landcover = ee.ImageCollection('USGS/NLCD_RELEASES/2016_REL').filter(ee.Filter.eq('system:index', '2016')).first().select('landcover')

roi = ee.FeatureCollection(ee.Geometry.Polygon(
    [[[-121.214213, 39.102459],
      [-121.214213, 39.109467],
      [-121.199295, 39.109467],
      [-121.199295, 39.102459],
      [-121.214213, 39.102459]]]))

points1 = landcover.sampleRegions(collection=roi, scale=30, geometries=True)

# I export this featureCollection as a CSV, compute a label for each point, and add the label to the points
# I assign a random label to demonstrate

def addLabel(point):
    return point.set({'label':random()})

points2 = points1.map(addLabel)

How can this updated featureCollection be visualized as an image with "label" as its pixel values?

Answer 1

Paint the points into an empty image:

ee.Image().double().paint(points, "label")