Generalized maps are widely used to model the topology of nD objects (such as images) by means of incidence and adjacency relationships between cells (vertices, edges, faces, volumes, ...). In this paper, we define a first error-tolerant distance measure for comparing generalized maps, which is an important issue for image processing and analysis. This distance measure is defined by means of the size of a largest common submap, in a similar way as a graph distance measure may be defined by means of the size of a largest common subgraph. We show that this distance measure is a metric, and we introduce a greedy randomized algorithm which allows us to efficiently compute an upper bound of it.