3D Digital Scanning Based On Binocular Vision And Research On Point Cloud Data Processing Technology

3D digital imaging is an emerging cross-high-tech field formed on the basis of integrated optics,microelectronics,computer science and other technologies.The 3D imaging technology based on binocular stereo vision utilizes two cameras or one camera at different positions to shoot the same scene through translation or rotation,and calculates the three-dimensional coordinate value of the point by calculating the parallax of the physical point in the two images.Thereby an image of the scene is obtained.This technology has become an important method in the field of 3D imaging due to its low price and wide applicability.The main research content of this paper is based on camera calibration technology,stereo vision calibration technology,3D point cloud display technology and point cloud data processing technology in binocular stereo vision technology.The following is the specific research content of this article:(1)The basic principle of binocular stereo vision,camera imaging principle,and the binocular stereo vision system model are analyzed on these theoretical principles.In this paper,the camera is calibrated by using the plane calibration target.The internal and external parameters of the binocular vision platform and the corresponding relationship between the three-dimensional point of the surface of the space object and the two-dimensional point of the image plane of the camera are obtained.(2)The phase demodulation algorithm and phase unwrapping algorithm based on 3D digital imaging technology are discussed,and the phase reconstruction process of the surface features of the measured object is realized.The phase demodulation and phase expansion algorithms suitable for this experiment are introduced and selected as the subject research.Focus on the time dimension phase reconstruction algorithm.(3)After projecting the stripe structured light on the measured object,the deformed fringe pattern carrying the height information of the object is collected from different angles by the camera,and the decoded deformed fringe pattern can obtain the absolute phase distribution corresponding to the depth image,and the corresponding points are matched and combined.The results of the previous calibration finally resulted in the results of the 3D point cloud data.(4)Discussed several point cloud data processing techniques.The final point cloud data is obtained by using the parallax data obtained by the four-step phase shift demodulation algorithm and the time phase unwrapping algorithm and combining the previous binocular stereo vision system calibration results,and the points obtained in different directions are obtained by manual stitching.The cloud data is spliced to form a complete point cloud data display result of the surface feature information of the measured object.(5)Several three-dimensional point cloud data processing technologies were completed:filtering,streamlining,sampling,encapsulation,and different algorithm effects and results comparison in different point cloud processing were analyzed.The optimal point cloud data processing method was selected and obtained.Better point cloud data,better to restore the surface information of the measured object.