Surface Quality Inspection Of Aluminum Alloy Weld Based On Line Structured Light

Welding technology is one of the most important material forming and processing technologies in modern manufacturing.Improper control of welding parameters can easily lead to the appearance of a series of weld surface defects,which directly affects the safety of welding products.Weld surface quality inspection is generally completed based on active vision technology.As a high-precision,non-contact sensing technology,this technology has become a hot spot in the study of weld surface quality inspection.In order to realize the online inspection of the surface quality of aluminum alloy welds,a visual inspection system based on linear structured light was built,and the composition and structure of the linear structured light sensor were discussed in detail.The imaging model is used to obtain the transformation relationship between the image coordinates of the weld feature points and the three-dimensional coordinates of the camera coordinate system.In actual image acquisition,it may be interfered by factors such as electronic noise and material reflection,which will affect the feature analysis of weld images.Therefore,image preprocessing operations such as projection positioning method,median filtering,and connected domain analysis are used to suppress the interference factors.;Secondly,according to the gray distribution of the laser stripe and the continuity of the center line,an improved geometric center method is proposed.This method advances along the interior of the center of the light strip,and combines the local OTSU adaptive threshold to achieve low brightness.The center line with good connectivity and smoothness is obtained by extracting the center of the uneven light strip.Taking butt weld,fillet weld and lap weld as the research objects,a maximum distance method is proposed.The feature point extraction algorithm combined with the recursive method,this method can accurately extract the weld feature points for different weld types,and the robustness of this method is verified by experiments;Finally,the weld seam is scanned,collected and counted,and the 3D reconstruction of the weld surface is realized.In order to accurately identify the types of weld surface defects,the characteristics of the surface defects of aluminum alloy welds were quantified according to the inspection standards,and the four types of welds with pores,undercuts,cracks and no defects on the weld surface were briefly characterized.Analysis,the DFS algorithm is used to classify the porosity and crack defects.The test results show that the detection accuracy of the system for the surface quality of the weld is as high as 97.43%,and the measurement accuracy in the defect area reaches 0.3mm.Based on the Qt framework and combined with multi-threading technology,the development and integration of the application program improves the utilization of system resources and the execution efficiency of the program,so as to realize the online inspection of the surface quality of aluminum alloy welds.