Font Size: a A A

Research On Localization Technology Of Steel Arch Based On Binocular Vision

Posted on:2022-05-29Degree:MasterType:Thesis
Country:ChinaCandidate:H HuangFull Text:PDF
GTID:2492306572980799Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
In the process of tunneling with rock tunneling machine,it is necessary to assemble the arc steel arches into annular structures in time to support the surrounding rock of the tunnel and prevent the occurrence of collapse accidents.At present,there are some problems such as low efficiency and high risk in manual aided steel arch assembly.Therefore,this paper studies the visual localization technology of steel arch frame,and develops a steel arch localization system based on binocular vision,which is combined with the manipulator to grab the steel arch frame for automatic assemblyThe main research contents of this paper are as follows:(1)Based on the analysis of the structural characteristics and assembly mode of the steel arch frame,a binocular vision localization scheme with hinged holes and edges as measurement features is designed.According to the requirement of steel arch localization precision,the hardware selection and integration of steel arch localization system are carried out.(2)Comparing with many image filtering algorithms,Gaussian filtering is chosen to filter steel arch image according to algorithm efficiency and detail retention ability.A multi-objective template matching algorithm based on mask is proposed,which can effectively detect ROI of the left and right hinged holes in the image of steel arches.An adaptive threshold algorithm based on area proportion was proposed to separate the hinged hole from the background inside the hole.Experiments show that the robustness of the algorithm is better than OTSU.(3)The error simulation of spatial linear reconstruction is carried out,and the influence of image linear feature extraction error on spatial linear reconstruction is analyzed,which provides guidance for improving the precision of steel arch edge measurement.An improved algorithm of spatial circle location based on spatial geometry is proposed,which solves the problem of double solution of spatial circle location and improves the practicability of spatial circle location algorithm.(4)The localization software of steel arch was developed,and the functions of image acquisition,feature extraction and pose calculation were realized by programming.The localization experiment of steel arch is designed to verify the accuracy of the localization algorithm.The experimental results show that the measurement accuracy of steel arch hinge hole features is ±0.1mm in image plane direction and ±0.2mm in depth direction.The measuring accuracy of steel arch edge intersection is ±0.03 mm,and the measuring accuracy of edge direction is within 1°,which meet the requirements of automatic steel arches stitching.
Keywords/Search Tags:Binocular vision, Image processing, Reconstruction of spatial primitives, Machine vision system, Localization of steel arch
PDF Full Text Request
Related items