Let's take a look at how an image can be visualized with the following code:

#import all required lib
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
from skimage.io import imread
from skimage.transform import resize

# Load a color image in grayscale
image = imread('sample_digit.png',as_grey=True)
image = resize(image,(28,28),mode='reflect')
print('This image is: ',type(image), 
         'with dimensions:', image.shape)

plt.imshow(image,cmap='gray')

We obtain the following image as a result:

def visualize_input(img, ax):

    ax.imshow(img, cmap='gray')
    width, height = img.shape
    thresh = img.max()/2.5
    for x in range(width):
        for y in range(height):
            ax.annotate(str(round(img[x][y],2)), xy=(y,x),
                        horizontalalignment='center',
                        verticalalignment='center',
                        color='white' if img[x][y]<thresh else 'black')

fig = plt.figure(figsize = (12,12)) 
ax = fig.add_subplot(111)
visualize_input(image, ax)

The following result is obtained:

In the previous chapter, we used an MLP-based approach to recognize images. There are two issues with that approach:

• It increases the number of parameters
• It only accepts vectors as input, that is, flattening a matrix to a vector

This means we must find a new way to process images, in which 2D information is not completely lost. CNNs address this issue. Furthermore, CNNs accept matrices as input. Convolutional layers preserve spatial structures. First, we define a convolution window, also called a filter, or kernel; then slide this over the image.