Depth Perception And 3D Reconstruction Based On Binocular Stereo Vision

With the rapid development of science and technology,computer vision technology plays an increasingly important role in the field of world cognition.As one research hotspot,binocular stereo vision technology is widely used in micro operating system,robot navigation,non-contact three-dimensional measurement and VR and other fields,so depth perception and three-dimensional reconstruction based on binocular stereo vision research has very important significance.This paper will focus on binocular stereo feature detection technology,matching algorithm,depth perception and three-dimensional reconstruction.The main contents are as follows:In the feature detection phase,the article analyzes the currently used corner detection algorithm and edge detection algorithm.For the purpose of our research,this article selects Harris corner extraction algorithm and make some optimization,the experiment result verifies that the optimized algorithm show some advantages in accuracy and anti-noise.Edge detection section highlights the typical differential operator and the optimal operator with the optimal filter optimal operator,and applies Canny edge detection algorithm in the experiment.In the stereo matching stage,we introduce the significance of polar-line correction and some basic concepts belong to stereo matching.we proposed an algorithm based on Census Transform and associated confidence plane fitting on the basic existing stereo matching algorithm,greatly reduces the mismatch rate about the matching window center pixel and improves the depth map's continuity and accuracy.Then we introduce the structural model of binocular stereo vision camera,describe the four present coordinates in details,and the coordinates of an arbitrary point in space from the image coordinate system to the world coordinate system are derived.Using a parallel model of binocular triangulation principle extracts depth information and carrying out the error analysis.Finally we discuss the Delaunay triangulation for spatial set and the characteristics of the Delaunay triangulation.Then we use the Lawson point by point insertion algorithm constructing triangular grid of the discrete spatial points,then we use OpenCV accomplishing the three-dimensional reconstruction of objects and use OpenGL complete the visualization of the reconstruction model.