The fingerprint ridge structure can be directly generated from its following features: the position of minutiae, the orientation field and the ridge frequency map. It was once believed that minutiae template does not reveal sufficient information of the orientation field and ridge frequency map to allow the reconstruction of the original fingerprint image. This belief has now been shown to be false; several algorithms have been proposed to generate the fingerprint ridge structure with uniform ridge frequency by reconstructing the orientation field from minutiae template. However, these reconstruction techniques assume equal ridge frequency in the whole foreground, which results in generating spurious minutiae along the changes of ridge curvature. In this paper, a novel algorithm for reconstructing the ridge frequency map from the minutiae template is proposed to complete the last piece of features required for the reconstruction of fingerprints. The change of local ridge frequency can be perfectly modeled by the minutiae and the divergence of the vectorial orientation field. The reconstructed ridge frequency map has been evaluated to be highly correlated to the ridge frequency map estimated from the original fingerprint images.