In this work, we develop a general method for estimating the Shannon entropy of a bidimensional sequence based on the extrapolation of block entropies. We apply this method to analyze the spatial configurations of cities of different cultures and regions of the world. Findings suggest that this approach can identify similarities between cities, generating accurate results for recognizing and classifying different urban morphologies. The hierarchical clustering analysis based on this metric also opens up new questions about the possibility that urban form can embody characteristics related to different cultural identities, historical processes, and geographical regions.