Convert numpy array of a image into blocks

Question

I have of a numpy array of a image.I want to convert this image into 8*8 block using python.How should I do this?

HYRY · Accepted Answer · 2018-02-21 02:56:30Z

2

reshape and then swapaxes:

import numpy as np

img = np.random.randint(0, 255, size=(128, 256, 3)).astype(np.uint8)

blocks = img.reshape(img.shape[0]//8, 8, img.shape[1]//8, 8, 3).swapaxes(1, 2)
print(blocks.shape)

to check the result:

np.allclose(blocks[0, 0], img[:8, :8, :])
np.allclose(blocks[3, 2], img[3*8:3*8+8, 2*8:2*8+8, :])

answered Feb 21, 2018 at 2:56

HYRY

97.8k28 gold badges197 silver badges192 bronze badges

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

Comments

SivaTP · Accepted Answer · 2018-02-21 02:36:46Z

1

Please provide your array structure.

you can use img_arrary.reshape(8,8), to work total elements must be 64

answered Feb 21, 2018 at 2:36

SivaTP

1532 silver badges7 bronze badges

2 Comments

Gayan Jeewantha Over a year ago

In reshape function only get the whole image shaped.But I want to access the image data block by block.How sholud I do this?EX: It is needed to seperate whole image as seperate 8*8 pixel blocks.

SivaTP Over a year ago

you can use slicing or boolean masking. Just look at on Numpy Slicing tutorial, you will get it easily.

Collectives™ on Stack Overflow

Convert numpy array of a image into blocks

2 Answers 2

Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related