Research And Design Of Large-size Workpiece Circular Hole Measurement System Based On Machine Vision

In the modern manufacturing industry,large-size workpieces face difficulties in automatic detection of surface defects and dimensions of large-size workpieces due to their wide width,long size and complex surface structure,and it is difficult to meet the requirements of full automation of production lines,especially for large-size workpieces.There is no good solution to the problem of high-precision automatic measurement of the dimensions of the geometrical features of circles and arcs.In this paper,taking the 3C product shell workpiece as the research object,a hardware platform for image acquisition of large-size workpieces based on machine vision is built,and two online size measurement methods for threaded holes on large-size 3C shell workpieces are proposed.The position size and hole size measurement of the threaded hole of the product shell workpiece.The main work of the paper includes:(1)A Zernike moment sub-pixel edge detection method based on improved Canny edge detection is proposed,and a reference coordinate system for dimensional measurement of geometric features of large-sized workpiece surfaces is established to achieve accurate positioning of workpiece contours.On the basis of the rough edge positioning of Canny detection,it can avoid the huge amount of calculation caused by the global integration of Zernike moments,improve the detection speed and achieve edge detection accuracy of 0.1-0.3pixel;(2)For 3C product shell workpieces,the corner positioning thread circle is used According to the characteristics of hole processing and the continuity of sub-images collected by line scan cameras,a method for measuring the position and size of threaded circular holes based on linear feature and circular feature positioning is proposed.The center coordinate extraction and diameter size measurement of the threaded hole are based on the continuity of the image,and the sub-images are detected one by one,and the starting pixels of adjacent sub-images are accumulated to realize the global detection of the size of the threaded hole in the large-size image.The method finally realizes that the position and size measurement accuracy of the threaded hole of the largesized 3C shell workpiece is ±0.2220 mm and the aperture size measurement accuracy is±0.0904 mm,and the single workpiece detection efficiency is 33.737s;(3)Based on the research(2),a new method is proposed.A method for measuring the size of the screw hole of the workpiece based on the prior position relationship of the circular hole.The method pre-locates the position of the threaded hole based on the size of the drawing,and decomposes the panoramic image of the workpiece into a region of interest for size information detection.This method solves the problem that the high-precision circle detection method requires global retrieval and mapping of high-resolution image edge information,which leads to system memory leakage and slow processing speed.The dimensional accuracy of the diameter of the threaded hole is ±0.0690 mm,and the detection efficiency of a single workpiece is 77.628 s.