Research On Parameter Detection System Of Rebar Head Based On Machine Vision

The rebar is an indispensable structural material in building components.In order to meet the quality standards requirement of building structures,it is required to detect the parameters of the rebar head with thread during production to ensure that the connect thread steel and the sleeve have a high mtching accuracy.In recent years,machine vision technology begun to be applied in the field of thread measurement.However,when detecting rebar heads,there are deficiencies such as inaccurate pitch diameter measurement,lack of angle detection of the end face of silk,and distortion of the head thread shield.Therefore,it is of great practical significance to study the parameter detection system apply to building rebar head with threads.Requirements for field testing of threaded steel heads in buildings are analyzed.This paper takes rebar as the research object,and machine vision technology is used to design a threaded wire head parameter detection system.By analyzing and calculating the tooth corners of the threaded wire head,the parameters of the threaded wire head are obtained.Firstly,according to the measurement requirements,the hardware platform equipment is selected,.Taking into account factors such as high noise in the process and the presence of coolant on the surface of the rebar,median filtering is used to achieve image noise reduction,Otsu is selected for image segmentation,and the Canny operator is used to obtain the image Edge.In order to reduce the influence of pseudo-corner points on the subsequent fitting,the method of variance curvature angle and gray area ratio is used to remove the pseudo-corner points,and on this basis,the sub-pixel subdivision algorithm based on Sigmoid edge is studied to achieve Edge corner subdivision.On the basis of corner sub-pixel subdivision,the fitting method of size,diameter,pitch and tooth shape angle is designed according to the definition of rebar parameters.In addition,for the problem of large error in the measurement method of the pitch diameter fitting,the adaptive nonlinear convergence whale algorithm is used to optimize the alculation method of the pitch diameter line of the thread with the smallest absolute value of the adjacent tooth line,which reduces the pitch diameter fitting error.In order to detect whether the end surface of the screw steel is flat,a method of angle fitting of the end surface of the rebar head with thread is designed.For the thread shape of the threaded wire head image caused by vertical illumination,there is a phenomenon of occlusion distortion,that leads to the error of pitch diameter measurement.The equations of the true tooth shape and the distorted tooth shape is derived,by which the diameter measurement error is compensated.Finally,a system software and hardware experiment platform was built on site,and the measured data was compared with the measured value of the standard instrument.Experimental results show that the system meets the requirements of on-site testing,which proves the effectiveness of the scheme.This subject realizes the parameter detection of the threaded steel wire head of the building,which guarantees the safety of the building structure to a certain extent,and has certain theoretical research significance and engineering application value.