Why is matplotlib of a circle producing color gradient along circle edges?

Question

I am using matplotlib (unsure as to the version) on python v2.7 to produce an image containing circles for use with an OOF2 analysis. The circles need to have pixels composed entirely of the same RGB values (#000000), however when the circles are generated, there is a ring of "greys" around the edge. coll_intra = matplotlib.collections.PatchCollection(circle_intra, facecolors='black', edgecolors='black')

Changing edgecolors between 'black' and 'none' has an effect, but does not set the values to be absolutely equal.
Circles with rings of "grey" around them:

Is there a way to adjust this in my script, or do I need to brute-force it in another program? (e.g. MSPaint).

Answer 1

Some grey colors can be present in the picture due to antialiasing.
One has to distinguish between the antialiasing that is performed by matplotlib and the one that may be present in the image viewer (e.g. the browser).

Concerning matplotlib, there is the option to turn antialiasing off by setting antialiased = False. Most artists and collections do have this option. So in this case

PatchCollection(...,antialiased=False)

would do the trick.

To observe the difference consider the following script. Setting antialiased to True (which is the default) outputs 121 different grey shades in the picture, while setting it to False leaves us with only 2 (black and white).

antialiased=True:

antialiased=False:

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.patches import Circle
from matplotlib.collections import PatchCollection

antialiased=False

N = 3
x = [1,2];y=[1,1]
radii = [0.5,.4]
patches = []
for x1, y1, r in zip(x, y, radii):
    patches.append(Circle((x1, y1), r))

coll_intra = PatchCollection(patches, facecolors='black', edgecolors='black',antialiased=antialiased)

fig, ax = plt.subplots(figsize=(2,1))
ax.set_aspect("equal")
ax.axis("off")
ax.set_xlim([0,3])
ax.set_ylim([0,2])
ax.add_collection(coll_intra)


#count the number of different colors
#https://stackoverflow.com/questions/7821518/matplotlib-save-plot-to-numpy-array
#https://stackoverflow.com/questions/40433211/how-can-i-get-the-pixel-colors-in-matplotlib
fig.canvas.draw()
data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep='')
data = data.reshape((int(len(data)/3), 3))
data = np.vstack({tuple(row) for row in data})

print len(data) # prints 121 for antialiased=True
                    #          2 for antialiased=False

plt.show()

(The method for counting the colors is adapted from this and this question.)