Objectives  Multi-view vertical line locus (MVLL) is a practical multi-baseline stereo image matching method, which can match and obtain the best elevation of ground points with the ground plumb line as geometric constraint. Aiming at solving the problem of independent ground point matching and lacking integrity constraint in MVLL matching method, the local smooth property of object space is applied to the matching process, and the MVLL matching method for multi-baseline stereo imagery based on semi-global constraint is proposed.

  Methods  Firstly, the accurate elevation of ground points is searched along the ground plumb line, which can be equivalent to the accurate parallax search along the epipolar image space. Secondly, the MVLL matching method is used to calculate the equivalent image matching measure of ground points on multiple images. Then, the semi⁃global matching (SGM) method is used to aggregate and analyze the matching measure through multi-path, and the equivalent disparity map under the local smooth constraint of the object space is obtained. Finally, the equivalent parallax map is converted to matching elevation of the ground points. And through multi-resolution matching of the multi-baseline stereo imagery, MVLL matching is realized under the overall optimal conditions and greatly integrated with SGM matching method.

  Results  The effectiveness of the proposed method is verified by experiments and analysis of various terrain features, including uneven surface feature, similar texture feature and occlusive feature. Comparative experiments are also conducted on local image areas.

  Conclusions  The experimental results show that the proposed method can optimize the object space matching measure of different terrain features, obtain more reliable matching results, and have higher image matching performance.