Author:
Wang Bin,Liu Lixia,Xue Wenjuan,Yang Fenglong,Hong Youxie,Zhou Yong,Wang Kai,Niu Hairui
Abstract
Abstract
Aiming at the problems of frequent low-level quality and difficulty in improving assembly efficiency in the assembly process of satellite honeycomb sandwich panel insert, the 3D model of the honeycomb sandwich panel is formed into high-definition model projection image by feature processing and extraction for 3D process information, and the virtual-real fusion insert aided assembly system is constructed, which is composed of core components such as camera, high-definition projector, and measurement software. Through structured light projection technology and projection algorithm based on landmark point transformation, the projection image can be accurately matched with the contour and hole position of the physical, structural panel, which can effectively assist the assembly process of the insert and reduce the low-level quality problems such as wrong use of parts and reverse polarity assembly. Compared with the traditional manual assembly method, the efficiency is improved by more than 200%, and the error rate is reduced by more than 300%.
Reference6 articles.
1. Machine vision intelligence for product defect inspection based deep learning and Hough transform;Wang;Journal of Manufacturing Systems,2019
2. FMCW signal detection and parameter extraction by cross wigner-Hough transform;Yasin;IEEE Trans. on Aerospace and Electronic Systems,2017
3. A method for registration of 3D shapes;Besl;IEEE Transactions on Pattern Analysis & Machine Intelligence,1992
4. Research on image processing method for measuring geometric parameters of contact line with binocular vision;Zhang;Journal of Qingdao University: Natural Science Edition,2016
5. 3D Object Recognition Based on Structured Light Projection;Song;Journal of optoectronics laser,2019