In this paper, we propose an efficient method to reconstruct the digital elevation model (DEM) map using the minimum spanning tree based on the congruence structure of the phase unwrapping problem. The proposed method consists of three main steps: First, we construct the system of congruence equations with smaller moduli based on the multichannel data. Next, we formulate and solve the congruence equation to obtain the optimal solution along with confidence coefficients. Finally, using these confidence coefficients as edge weights, we apply Kruskal’s algorithm to construct the minimum spanning tree and reconstruct the DEM map based on the connectivity of its endpoints. Experiment results demonstrate that the extended method outperforms both the RRNS method and the minimum cost flow method.