如何在python中将图像划分为块？

Question

I am using keras and I want to read an image in my code and then divide it into 32x32 non-overlapping blocks and send each block into my learned deep learning model for further processing. how can I do it? during learning, I used CIFAR10 for training which has images with size 32x32. now I want to load Lena and then divide it into 32x32 blocks and send to my model for processing. do I need to do some changes on Lena to achieve correct accuracy? Thank you.

lena=Image.open('E:/images/lena_gray_512.tif')
a=view_as_blocks(lena_part, (32,32))

error:

a=view_as_blocks(lena_part, (32,32)) Traceback (most recent call last):

File "", line 1, in a=view_as_blocks(lena_part, (32,32))

File "D:\\software\\Anaconda3\\envs\\py36\\lib\\site-packages\\skimage\\util\\shape.py", line 82, in view_as_blocks if block_shape.size != arr_in.ndim:

AttributeError: 'TiffImageFile' object has no attribute 'ndim'

Answer 1

You can try view_as_blocks

from skimage.util.shape import view_as_blocks

view_as_blocks(arr, (2,2))