Drawing circles on image with Matplotlib and NumPy

Question

I have NumPy arrays which hold circle centers.

import matplotlib.pylab as plt
import numpy as np
npX = np.asarray(X)
npY = np.asarray(Y)
plt.imshow(img)
// TO-DO
plt.show()

How can I show circles at the given positions on my image?

Possible duplicate of [plot a circle with pyplot](http://stackoverflow.com/questions/9215658/plot-a-circle-with-pyplot) — kazemakase, Jan 20 '16 at 14:24
Quite so. Answers to that question show how to draw circles, which is exactly what you asked for :) — kazemakase, Jan 20 '16 at 14:28
If you want to draw circles directly onto a numpy array, you can use the Python Imaging Library. See my answer at http://stackoverflow.com/questions/12638790/drawing-a-rectangle-inside-a-2d-numpy-array; change `draw.polygon(...)` to `draw.ellipse(...)`. See the PIL docs for details: http://effbot.org/imagingbook/imagedraw.htm — Warren Weckesser, Jan 20 '16 at 14:55

score 33 · Accepted Answer · answered Jan 20 '16 at 16:36

33

You can do this with the matplotlib.patches.Circle patch.

For your example, we need to loop through the X and Y arrays, and then create a circle patch for each coordinate.

Here's an example placing circles on top of an image (from the matplotlib.cbook)

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

# Get an example image
import matplotlib.cbook as cbook
image_file = cbook.get_sample_data('grace_hopper.png')
img = plt.imread(image_file)

# Make some example data
x = np.random.rand(5)*img.shape[1]
y = np.random.rand(5)*img.shape[0]

# Create a figure. Equal aspect so circles look circular
fig,ax = plt.subplots(1)
ax.set_aspect('equal')

# Show the image
ax.imshow(img)

# Now, loop through coord arrays, and create a circle at each x,y pair
for xx,yy in zip(x,y):
    circ = Circle((xx,yy),50)
    ax.add_patch(circ)

# Show the image
plt.show()

answered Jan 20 '16 at 16:36

tmdavison

50,222
11
138
131

1

it is circles in plot, not on the image. img would not be changed – Andrew Matuk Mar 03 '17 at 20:46
@Andrew one can save the image by using plt.savefig('grace_hopper.png') – user1953366 Mar 19 '21 at 04:51
@user1953366 saving to disc is never a good idea, there are many drawbacks - it would be very slow and cause out of disc space if you want to make it millions of times. Also, you assume that image encoding is lossless. – Andrew Matuk Mar 21 '21 at 11:11
Plz see my other answer (formatting is not correct here) – user1953366 Mar 21 '21 at 13:48

score 1 · Answer 2 · answered Mar 21 '21 at 13:51

To get the image, instead of plt.show do (Without saving to disc one can get it as):

io_buf = io.BytesIO()
fig.savefig(io_buf, format='raw')#dpi=36)#DPI)
io_buf.seek(0)
img_arr = np.reshape(np.frombuffer(io_buf.getvalue(), dtype=np.uint8),
                         newshape=(int(fig.bbox.bounds[3]), int(fig.bbox.bounds[2]), -1))
io_buf.close()
plt.close()  #To not display the image

Drawing circles on image with Matplotlib and NumPy

2 Answers2

Linked