3D Reconstruction for Robot Navigation Based on Projection of Virtual Height Line and its Performance Evaluation

3D reconstruction is a challenging work for the autonomous navigation of robots. This paper proposes a novel 3D reconstruction method based on projection of virtual height line (PVHL). At first partition the scene scope uniformly into grids with a certain size. Then introduce virtual height lines (VHLs) which do not exist in the scene. Each VHL is projected on left image and right image, and two projection lines are generated. Each projection line has a point of intersection with the scene, which is the only existent and visible point on the projection line. So this point has the maximum similarity metric, and the height of the point is that of the corresponding scene. Finding the height of the scene is converted to prove whether the point of intersection has the maximum similarity metric or not. The experiments to reconstruct 3D scene are performed and the results show the proposed method is feasible and effective.