The area of map merging is tightly connected to the area of image processing. Usually metric maps created by robots are represented as occupancy grids. It is easy to apply algorithms used in image processing to this kind of map representation. The image processing subfield that is closest to the map merging is the image registration. It can be assumed that metric map merging methods similarly to the image registration methods consist of three components: feature space, search strategy and similarity metric. Algorithms from image processing can also be used in map merging for map preprocessing. The goal of this paper is to explore similarities between the fields of map merging and image processing and to determine how the results of this research can be used for the development of a map merging framework and consequently new map merging approaches.