The Research On Obtaining Three-dimensional Data Of The Object By A Single Video Camcra

Obtaining the three-dimensional data of the object has a wide range of applications in themultimedia technology, reverse engineering, virtual reality and3D game production. And it isa key and difficult problem in the areas such as computer vision, artificial intelligence, virtualreality and so on. It is also one of the major challenges in basic research and applied researchfor the humanity.3D reconstruction and projection process is the inverse process.3D reconstructiondetermines a spatial point through a projection point. Considering that the points onattachment of a projection point and a camera attachment point of focus all meet theprojection equation, we can solve space points if we know two images. Currently, somemethods of3D reconstruction are facing some shortcomings, especially a great of mismatchpoints and systematic error in3D reconstruction.This subject use the stereo binocular parallax method to construct a set of3Dconstruction algorithm, then construct a3D reconstruction evaluation system to evaluate theaccuracy of the3D reconstruction algorithm from two aspects. On one hand, assume thatmatching feature points are exactly right; On the other hand, introduce error in matchingalgorithm and camera parameters calibration. Then use3D reconstruction algorithm toforecast the3D data respectively. Then we will compare the predicted3D data with theoriginal model to calculating the deviation, and concluded the accuracy of3D reconstructionalgorithm, and the passed error of matching algorithm and camera parameters calibration errorof produce in the process of3D reconstruction.The study found that reconstruction error will increase if we increase the matchingerrors; reconstruction error will be reduced if we improve Image resolution; Images of thesame increase displacement of the plane of the point that, by matching the error caused byerror transfer decreased; the passed error will decreased when we increase the movement ofthe image for the points in the same plane, and it is opposite for the points which are not inthe same plane because of the depth scale.