Research On Online Inspection For The Wall Thickness Of The Steel Pipe Based On Machine Vision

At present,manual sampling is often used to measure the wall thickness of steel pipe end face,which leads to low measurement efficiency,low precision and few measurement data.In this paper,the machine vision measurement technology was applied to the dimension measurement of workpiece,Taking the oil country tubular goods(OCTG)as the research object,this paper illustrates the research on the method of online detection of steel pipe end wall thickness based on the machine vision.The main research contents are as follows.(1)Hardware platform of wall thickness detection and systemic software module was designed.The design of the systemic hardware platform includes the design of the lighting system,the selection of industrial cameras and optical lens and so on,At the same time,the systemic process was explained,programming language C++ was used to complete the design of the detection algorithm based on VS2017.(2)The algorithm of image preprocessing and edge detection was carried out.The relevant preprocessing algorithm and edge detection algorithm had been analyzed,and the image preprocessing algorithm and edge detection algorithm conforming to the detection system in this paper were selected through the experimental test.(3)A kind of improved RHT circle detection algorithm was proposed.The RHT circle detection algorithm had been improved by analyzing the shortcomings of circle detection algorithm.The details are as follows: Obtaining three edge points that are noncollinear by random sampling from different area,then constructing the candidate circle;Selecting the candidate circle that fits the edge profile the most as the final detection circle.Comparing the improved RHT circle detection algorithm with Least Square Method circle detection algorithm,random sample consensus circle detection algorithm and RHT circle detection algorithm,it is approved that the circle detection algorithm of this paper can detect the circular outline more accurately and has strong robustness.(4)Software of steel pipe wall thickness detection system was designed.This paper illustrated the design of the host computer system on the wall thickness detection based on the QT framework,which can achieve system calibration,image preprocessing,circle detection,dimensional measurement and many other operations,and the database of steel pipe wall thickness has been designed based on My SQL,Using the detection system in this paper to detect the end of wall thickness of the steel pipe shows that this software can reach high detection accuracy,with measuring error of about 0.1mm.Based on proposed the research of an improved RHT circle detection algorithm,this paper designed the thickness of steel pipe end wall detection system online based on the machine vision,which can accurately measure the wall thickness all over the end of steel pipe,It has been proved that the detection system can basically meet the requirements of the detection accuracy and speed of the wall thickness detection through the experiments,and has certain significance in project.