Font Size: a A A

Technique Research For Sheet Metal Edge Inspection Based On Binocular Vision

Posted on:2021-10-03Degree:MasterType:Thesis
Country:ChinaCandidate:Z Y LiuFull Text:PDF
GTID:2481306479458404Subject:Aviation Aerospace Manufacturing Engineering
Abstract/Summary:PDF Full Text Request
Sheet metal parts are one of the most common parts in the aerospace industry,and are widely used in aircraft aerodynamic shapes,body frames and interior decoration structures.The inner / outer contour edges on a sheet metal part are important information to characterize the shape of the sheet metal part,which is critical for the manufacture and assembly of sheet metal parts.Aiming at the difficult problem of rapid and accurate inspection of sheet metal edges,research on edge inspection technology of sheet metal parts based on binocular vision is carried out.The main research contents are as follows:1.A complete set of edge detection technology solutions for sheet metal parts based on binocular vision is proposed.Based on this solution,a set of edge detection systems for sheet metal parts based on binocular vision(SMEIS)is designed and developed,including measurement sensor hardware devices and systems supporting software.The hand-held inspection device is continuously moved around the edge of the measured sheet metal with the round marks on the surface,while the line laser transmitters project laser stripes to the edge of the sheet metal part.The cameras acquire the synchronized images in real time and transmit them to the computer.The system software performs incremental 3D reconstruction of the laser stripe centers on the sheet metal edge in real time via parallel processing.2.A set of real-time 3D reconstruction algorithms for laser stripe center points are proposed.The core of the algorithm mainly includes real-time extraction algorithms for laser stripe center points based on ROI,and laser stripe center points matching algorithms based on epipolar constraints and line fitting.3.A set of real-time registration algorithms based on circular marks are proposed.The core of the algorithm consists of coarse and fine extraction of the center of the circular marks,a matching algorithm of the center of the round marks based on the disparity gradient constraint,a real-time registration algorithm of the point cloud based on the Euclidean distance triangle,and a local and global optimization algorithm based on bundle adjustment.4.The edge inspection experiments of sheet metal were carried out using SMEIS.The experiment results indicate that the developed inspection system has high real-time performance and precision.
Keywords/Search Tags:Sheet metal edge, Binocular vision, Line laser, Real-time inspection, Three-dimensional reconstruction
PDF Full Text Request
Related items