Algorithm 2 Learning algorithm for CNN model.

Input: a grayscale image, represented as a single-channel image. Parameter: N_epochs, Batch_size, SGD optimizer, Cross-entropy loss function. Output: the probabilities of the image belonging to each of the classes. 1.Initialize the neural network with a specific architecture (layers and connections) and initial weights. 2.Load the data and set up the training parameters. 3.for in N_epochs do 4.      Set the neural network in training mode. 5.      for Batch_size in training dataset do 6.          Transfer the data and target values to the device (GPU or CPU) based on the settings. 7.          Reset the gradients of the optimizer. 8.          Perform a forward pass through the model to obtain the output values. 9.          Calculate the loss function between the predicted and target values. 10.           Perform backpropagation of gradients to compute them for each model parameter. 11.           Update the model weights using the SGD optimizer. 12.     end for 13.end for 14.Finish the algorithm execution.