This paper investigates the use of graph cuts for the minimization of an energy functional for road detection in satellite images, defined on the Bayesian MRF framework. The road identification process is modeled as a search for the optimal binary labeling of the nodes of a graph, representing a set of detected segments and possible connections among them. The optimal labeling corresponds to the configuration that minimizes an energy functional derived from a MRF probabilistic model, that introduces contextual knowledge about the shape of roads. We formulate an energy function modeling the interactions between road segments, while satisfying the regularity conditions required by the graph cuts based minimization. The obtained results show a noticeable improvement in terms of processing time, while achieving good results.