Plot 2 dimensional numpy array in .npy file using matplotlib?

Question

I have numpy file stored as "test.npy" which is a 2 dimensional Synthetic Aperture Radar(SAR) image data with VV and VH polarization bands. How do I plot this 2 dimensional image array using matplotlib?

import numpy as np

img_array = np.load('test.npy')
from matplotlib import pyplot as plt

plt.imshow(img_array[1], cmap='gray')
plt.show()

But the line:

plt.imshow(img_array[0], cmap='gray')

plots only the first band in the list. So, how is it possible for me to plot 2 dimensional image array?

have you tried plt.imshow(img_array)? You can pass the 2-d numpy array directly in to imshow() — NMme
– NMme, Commented Aug 16, 2021 at 13:36
yeah I tried that and got an error like this one: "TypeError: Invalid shape (2, 512, 512) for image data " — hillsonghimire
– hillsonghimire, Commented Aug 16, 2021 at 13:52
Also, "print(img_array.shape)" returns the shape of image array as: (2, 512, 512) while "print(img_array[0].shape" returns the shape of the image array as: (512, 512). Does this have to do anything with the original shape of my img_array as (2, 512, 512). — hillsonghimire
– hillsonghimire, Commented Aug 16, 2021 at 13:58
Please include a minimal reproducible example of the data as formatted text — Trenton McKinney
– Trenton McKinney, Commented Aug 16, 2021 at 16:02

NMme · Accepted Answer · 2021-08-16 16:09:56Z

0

The problem is that your array has the shape (2, 512, 512) but matplotlib can only show grayscale, rgb or rgba images. Also the array needs to have the form (H,W), (H,W,3) or (H,W,4) respectively as you can see in the documentation here. Therefore I suggest you swap the axes and extend the array to a (H,W,3) shape with an additionally empty channel:

import numpy as np
from matplotlib import pyplot as plt

img_array = np.load('test.npy')
c, h, w = img_array.shape
img = np.zeros((h, w, 3))  # create image of zeros with (height, width, channel)
img[:, :, :2] = img_array.swapaxes(0,2) # fill the first 2 channels
plt.imshow(img)
plt.show()

edited Aug 16, 2021 at 16:09

answered Aug 16, 2021 at 14:10

NMme

4713 silver badges14 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

hillsonghimire Over a year ago

I still get this error: TypeError: Invalid shape (512, 512, 2) for image data

Ioannis Nasios · Accepted Answer · 2021-08-17 07:19:24Z

0

import numpy as np
from matplotlib import pyplot as plt

img2d = np.load('test.npy')
img3d = np.concatenate(( np.expand_dims(img2d[0],-1), np.expand_dims(img2d[1],-1),np.expand_dims((img2d[0]+img2d[1])/2, -1)), -1)


plt.imshow(img3d)
plt.show()

answered Aug 17, 2021 at 7:19

Ioannis Nasios

8,5474 gold badges41 silver badges59 bronze badges

Collectives™ on Stack Overflow

Plot 2 dimensional numpy array in .npy file using matplotlib?

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related