[FIXED] How to convert a matplotlib figure to a cv2 image?

Issue

I would like to know how convert a matplotlib figure to a cv2 image, this is my python3 code:

import cv2
import matplotlib.pyplot as plt
  
# Generating matplotlib figure and save it as a png image
plt.plot([0, 5], [0, 5])
plt.savefig('plot.png') 

# Reading figure saved as png image
img_plot = cv2.imread("plot.png")

# Displaying image using opencv imshow
cv2.imshow('Image', img_plot) 

# Waiting for any key to terminate image window
cv2.waitKey(0)

and the result after executing it:

Is it possible to display the figure without having to save it first?, something like this:

import cv2
import matplotlib.pyplot as plt
  
# Generating matplotlib plot
img_plot = plt.plot([0, 5], [0, 5])

# Displaying image using opencv imshow
cv2.imshow('Image', img_plot) 

# Waiting for any key to terminate image window
cv2.waitKey(0)

but this leads to an error.

I know it's possible to display the image with plt.show() as well, but I would like to use cv2.imshow

Solution

You can display a Matplotlib figure in OpenCV without saving it first. You need to convert the Matplotlib figure to a NumPy array before using cv2.imshow().

Below is an example:

import cv2
import matplotlib.pyplot as plt
import numpy as np

fig, ax = plt.subplots()
ax.plot([0, 5], [0, 5])

fig.canvas.draw()
img_plot = np.array(fig.canvas.renderer.buffer_rgba())

cv2.imshow('Image', cv2.cvtColor(img_plot, cv2.COLOR_RGBA2BGR))

cv2.waitKey(0)

Answered By - Jake Zappin