Research On Measurement Technology Of Brinell Hardness Indentation Based On Machine Vision

Hardness is a performance index to measure the hardness and softness of materials.To some extent,it reflects the ability of materials to resist elastic deformation,plastic deformation or damage.With more and more attention paid to modern industry in China,as the most important transition product in modern industry,the measurement method of hardness of metal materials is becoming more and more important.Brinell hardness measurement is one of the most widely used hardness measurement methods.The traditional Brinell hardness measurement is that operators use optical microscope to read data directly,which is easy to cause the operator’s eye fatigue and affect the measurement accuracy,but also reduce the measurement efficiency.Because of the non-contact measurement of machine vision,it can improve the degree of automation,and observe the same object for a long time Therefore,this paper takes Brinell hardness indentation as the measurement object,and studies the machine vision measurement technology in depth:(1)Aiming at the problem of camera radial distortion center positioning error,a camera radial distortion center correction method based on the normalized 8-point algorithm is proposed.In this method,the distortion center is found according to the corresponding relationship between the checkerboard and the matching corner of the checkerboard image,and the distortion center is determined by the normalized 8-point algorithm and the minimum two multiplication to realize the camera radial distortion center correction and eliminate the distortion The influence of the center on the camera calibration process improves the accuracy of the camera image distortion correction;(2)To solve the problem of incomplete edge contour extraction,an edge detection technology based on ant colony algorithm is proposed.This technology introduces ant colony algorithm on the basis of traditional edge detection algorithm to optimize the edge contour extracted by traditional edge detection algorithm,which makes the edge contour of indentation more complete and improves the accuracy of feature extraction;(3)In order to solve the problem of two edge profiles in the process of edge detection,after analyzing the section of Brinell hardness indentation,the principle of Brinell hardness test and the calculation formula of Brinell hardness value,the edge profile extracted from the boundary part of the indentation is selected as the edge profile of the indentation;To solve the problem of incomplete edge profile extracted from the indentation,an ant colony algorithm is proposed Based on the traditional edge detection algorithm,ant colony algorithm is introduced to optimize the indentation edge contour extracted by the traditional edge detection algorithm,which makes the indentation edge contour more complete and improves the accuracy of feature extraction;(4)Aiming at the Brinell hardness indentation which is studied in this paper,a new measurement method of Brinell hardness indentation based on machine vision is proposed by combining the correction method of camera radial distortion center based on normalized 8-point algorithm,edge detection technology based on ant colony algorithm and circular arc fitting detection algorithm based on least square method.