Research On Binocular Vision Laser Compensation 3D Reconstruction System

In recent years,computer vision has gradually become a research field for more and more people.The vision-based 3D reconstruction technology has bright application prospects due to its advantages such as no direct contact with the object,no damage to the object table,fast 3D reconstruction speed,and high efficiency.Among all threedimensional reconstruction technologies,the three-dimensional reconstruction of binocular vision is one of the ways to realize the three-dimensional reconstruction.It uses the left and right cameras at different angles and positions to reconstruct the image of the target object,or use a single camera to collect the image of the target object at two or more angles and positions,and then obtain the target object through a series of computer algorithms The parallax in each photo finally gets the depth data map,and then the 3D point cloud data is obtained from the depth data map.Although the 3D reconstruction technology based on binocular vision is becoming more and more perfect and is also beginning to be used in all walks of life,it still has some shortcomings so far,such as low reconstruction accuracy,poor reconstruction effect,and poor performance on target objects.The reconstructed point cloud data is prone to loss of some point cloud data.For this reason,this paper proposes a set of dot matrix laser ranging device to construct the 3D reconstruction system based on binocular vision laser compensation to solve the shortcomings of the above-mentioned binocular vision 3D reconstruction.This topic will design the 3D reconstruction device based on binocular vision laser compensation based on the principle of 3D reconstruction based on binocular vision.The main contents of this paper are as follows:First,in this subject,we built a dot matrix laser ranging device on the periphery of the camera on the basis of two cameras and light sources placed in different positions to obtain a point cloud reconstructed from the binocular vision.Correction coefficients for data restoration.This set of correction coefficients can optimize the initially obtained point cloud data to achieve a more accurate 3D model reconstructed and closer to the 3D object being reconstructed in the real world.Second,use the dot matrix laser ranging device built in this subject to repair the missing point cloud data and assist the final 3D model generated on the computer to be more complete.Third,according to the content in this article,we built a complete set of 3D reconstruction system based on binocular vision laser compensation.The program uses opencv3.1 to cooperate with MATLAB2014 for development and debugging under the Windows10 system environment.Through the final experiment,we proved that the device can quickly,completely and accurately reconstruct the target object in three dimensions,and its accuracy is sufficient to meet the daily production needs,and the reconstructed model is complete,and there is basically no loss of point cloud data.