- Binarization with the Otsu method (implementation from scratch – do not call the matlab function) (2 point)
- Binarization with Sauvola (1 point)
- Binarization with Niblack (1 point)
- Using the k-means algorithm, create images with k different colors ( k different shades of gray) (2 point)
- Using as features for each point: a neighborhood of size w x w (all the neighborhood is used) (1 point)
- Using as features for each point: the mean and standard deviation of a neighborhood of size w x w (1 point)
- Normalize the images so the image is a square (number of columns = number of rows), with the default background being black, and the gravity center of the image is placed in the middle of the image (1 point)
- Filter the images with Gabor filters by
