In this project I was asked to train a neural network to restore missing parts of an image. This was part of the Course Programming in Python 2 offered in the Johannes Kepler University Linz by Micheal Widrich.
The dataset used unfortunatly isn't public, however these are the requirements for the input:
image_array: A numpy array of shape (X, Y) and arbitrary datatype, which contains the image data.
crop_size: A tuple containing 2 odd int values. These two values specify the lengths of the rectangle that should be cropped-out in pixels for the two spatial dimensions X and Y in that order.
crop_center: A tuple containing 2 int values. These two values are the position of the center of the to-be cropped-out rectangle in pixels for the two spatial dimensions X and Y in that order.
