将图像从块转换为原始大小

Question

您好，我上传了一张256*256的图像，并按照以下步骤将其转换为8乘8的块

import numpy as np
import matplotlib.pyplot as plt
from scipy.io import loadmat
from PIL import Image
import scipy.misc as smp
from random import random
from math import exp

image = Image.open('jama.png').convert('L')

#image.show()
arrimag = np.array(image)
#print(arrimag.shape)
#print(arrimag)
row_windowsize = 8
column_windowsize = 8
imagelen = len(arrimag)
f = np.array([])
for row in range(0, imagelen, row_windowsize):
    for column in range(0, imagelen, column_windowsize):
        win = arrimag[row:row+row_windowsize, 
column:column+column_windowsize]
        window1 = np.zeros((64,1))
        window1 = win.ravel()
        window1 = window1.reshape(64,1)
        f = np.append(f,window1)
window = f.reshape((1024,64))
#print(window) 

所以，如果有人能帮我把代码倒转过来，这样我就可以把我的图像从（8*8）块恢复到原来的（256*256）块，我将不胜感激

注意：我的图像的新大小是Window（1024,64），我需要转换回Window（256256）