Research On3D Reconstruction Based On Binocular Stereo Vision Using The Ordinary Cameras

The technique about3D reconstruction based on binocular stereo vision imitates the human vision system. It uses two cameras, focusing on the same scene, to restore the depth information about the scene. There are two main topics about the technique:one is about camera calibration and the other is about stereo matching. And the purpose of the camera calibration is to obtain the accurate camera extrinsic and intrinsic parameters quickly. Meanwhile, the purpose of the stereo matching is to find the points in the stereo image pair quickly, which shoots on the same scene’s point by two cameras.With3D application developing and ordinary cameras widely using, we use the ordinary cameras as the experimental equipment. Concerning about the camera calibration technique, in this paper we uses the Zhang’s method as it is very convenient and practical. Meanwhile, we use the circle pattern template as the object for reference, and it can improve calibration precision. In regard to the stereo matching technique, in this paper we propose present a novel propagation-based stereo matching algorithm, Firstly, we select highly reliable depth points (i.e., support points) from both the high-textured areas and the low-textured ones in the rectified images. Then we propagate these support points along adjacent neighboring structure to produce the depths of other points in the image based on the triangulation derived from a set of highly reliable support points. This allows for efficient exploitation of the disparity search space, yielding accurate dense disparity without the need for global optimization.Finally, experimental results demonstrate both the effectiveness and convenience of the camera calibration method we used, at the same time, the experimental results also shows both the effectiveness and matching accuracy of the stereo matching method we proposed.