A recovering technique of an elevation map by stereo modeling of the real aerial image sequence is presented. An area-based stereo matching method is proposed and parameter values are experimentally chosen. In a depth extraction step, the depth is determined by solving a vector equation suitable for stereo modeling of aerial images that do not satisfy the epipolar constraint. Also, the performance of the conventional feature-based method is compared via computer simulations. Finally, techniques analyzing the accuracy of the recovered elevation map (REM) are discussed. The experimental results based on error performance show the efficiency of the proposed method.