Research On 3d Reconstruction Technology Based On Color Structured Light Coding

Structured light projection technology is a widely studied topic in the field of fast and low cost 3D imaging.Non-contact 3D surface profiling based on structured light is the preferred method for reverse engineering,industrial quality control of product parts,object recognition,medical and healthcare applications and other fields,because it does not destroy the original appearance of the object,can accurately reconstruct the original appearance of the object.One of the main challenges is to increase the density of the reconstruction while projecting as few patterns as possible,thus reducing the acquisition time for projected patterns.Based on color structured light coding principle,in this paper,the key technologies of color structured light are studied,put forward a kind of based on hamming distance of six yuan third-order De Bruijn sequences encoding,and color structured light coding and combination of binocular stereo vision,solved the surface texture information in binocular stereo vision is less object feature matching difficult problem,At the same time,a dense reconstruction of 3D information on projection is realized.The main research content of the thesis mainly has the following points:1.in view of the low texture object,based on the binocular stereo vision and the respective advantages and disadvantages of color structured light technology,combine the two,design based on binocular vision for the color structured light three-dimensional reconstruction program,and completed the cameras,projectors,slide rail stents and other hardware equipment selection and the construction of the system platform.2.The coding technology of color structured light is studied.Due to the influence of environmental light,object albedo,uneven channel gain and color crosstalk,it is a difficult problem to use color structured light to perform dense 3D reconstruction.In this paper,we propose a new color structured light coding based on Hamming distance,which can achieve a densier reconstruction effect than traditional color coding.An image enhancement algorithm based on fusion of RGB and HSV color space is proposed to solve the problem of color deviation in the obtained color stripe image.3.The decoding technology of color structured light is studied.When color structured light is used,color clustering in HSV space is adopted to identify the fringe color in order to ensure that the projected color can be correctly recognized from the data recorded by the camera.In order to extract the stripe ridges of the image,the gray threshold algorithm based on subpixel is used to extract the stripe ridges of the color structured light.Compared with the traditional center line extraction method,the center line of the color stripe can be extracted smoothly and correctly.For r the fringe matching of left and right images,the window unique matching strategy of De Bruijn sequence is used to perform fringe matching,and then the parallax information of left and right images is obtained.4.According to the reconstruction principle of binocular vision and color structured light,the 3D point cloud information of the object is calculated,and the point cloud information is visualized finally.Through three-dimensional reconstruction of different models and precision analysis,the efficiency and superiority of the experimental scheme in this paper are verified..