The tophat transform is a well know and widely used transform based of morphological opening or closing. It is often used to remove the background in an image, where the background is defined as the zones which do not fit in the user provided kernel.

Similarly, the tophat by area is used to remove the background in an image, but the criteria of selection is the size of the connected component, either in physical size or in number of pixels, and is independent of the shape of the connected component. This property makes the tophat non dependent of the shape of the object and makes easier to describe the criteria.