For the first processing steps after capturing a digital image, we need two classes of operations, point operations and geometric operations. These two types essentially modify the “what” and the “where” of an image point. They neither extract specific information from the image nor do they perform operations to recognize objects.