RGB image display in Matplotlib: plt.imshow returns a blue image

Question

I am trying to display an RGB image using matplotlib. This is the code:

# import libraries
import numpy as np
import cv2
from matplotlib import pyplot as plt

# use opencv to load the image
image = cv2.imread("path/to/file/image.jpg", 1)

# convert it to numpy array
pixels = np.array(image)

And then, when I try to visualize the image with:

plt.imshow(pixels)
plt.show()

It returns a picture that is all blue. I don't understand why, since the image is a normal colored image. I tried with multiple images, and the problem persists. Moreover, with my laptop at work I didn't get any problem.

score 35 · Accepted Answer · answered Mar 02 '19 at 14:32

35

You are facing this issue since you are reading the image with opencv and opencv reads and displays an image as BGR format instead of RGB color format. Whereas matplotlib uses RGB color format to display image. Try using:

image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

pixels = np.array(image)

plt.imshow(pixels)
plt.show()

answered Mar 02 '19 at 14:32

Apoorv Mishra

1,489
1
8
12

1

"pixels = np.array(image)" why do we need this ? thanks ! – Dipanwita Mallick May 24 '21 at 05:25
1

This is not necessary ,do not worry. – devansh Dec 15 '21 at 11:19
No need for cv2 dependencies. Just reverse array dimensions with `plt.imshow(img[:,:,::-1])`. See this example from ML models doing tree counting: https://deepforest.readthedocs.io/en/latest/getting_started.html – Maciej Skorski Aug 03 '23 at 11:08

score 29 · Answer 2 · edited Mar 06 '19 at 12:31

29

you can directly use this line to convert the BGR image to RGB:

image = cv2.imread("path/to/file/image.jpg")[:,:,::-1]

edited Mar 06 '19 at 12:31

noiaverbale

1,550
13
27

answered Mar 06 '19 at 10:41

Akhil prasannan

395
4
11

RGB image display in Matplotlib: plt.imshow returns a blue image

2 Answers2