Research On Piston Surface Inspection Technology Based On Machine Vision

With the rapid development of industrial manufacturing in our country,there are higher requirements for product measurement.In the field of intelligent manufacturing,the accuracy and efficiency of detection are greatly improved by using machine vision in the field of intelligent manufacturing.Therefore,the study of piston surface inspection technology based on machine vision is of great significance in the field of intelligent manufacturing.In order to solve the problems of low efficiency and high labor cost of piston inspection in the past,this paper designs a piston inspection platform,deeply studies the key problems and detection algorithms of machine vision piston inspection,and uses C # language to realize the interface of piston size detection system.The specific research contents are as follows:(1)This article introduces the piston detection hardware system,including the construction of the piston detection platform,the design of the visual measurement system,and the design of the control part.The hardware system design starts from the requirements of the various detection items of the piston,and integrates the corresponding detection methods of each item to form a detection platform that can complete all the detection items of the piston;according to the detection principle,a reasonable visual measurement system is designed and the motion control part is built.Lay the hardware foundation for piston detection.(2)The calibration algorithm of binocular vision system is studied in this paper.based on the traditional binocular camera calibration method,an improved calibration algorithm for calculating the spatial position relationship of binocular cameras is proposed.the calibration operation steps of the calibration method are designed,and the micron raster ruler and a small chessboard calibration board are skillfully used to obtain the camera parameters.Compared with the traditional measurement method,this method has obvious advantages in accuracy.(3)The detection algorithm of piston diameter,which requires high precision,is studied in this paper.when the edge of the target is not clear,an image reconstruction method to reduce the image and an improved edge detection operator are proposed.the problem of image distortion caused by optical diffraction is solved,the interference information in the image is reduced,and an error caused by mechanical problems is analyzed,and an error correction algorithm is proposed.The overall testing accuracy of the piston diameter is improved.(4)The method of measuring the size of the inner edge of the piston pin hole and the inner circle of the upper end of the piston is studied in this paper.The flow of the algorithm for detecting the inner edge size of the piston pin hole is proposed,and a target edge extraction method is designed for the redundant edges in the detection process.The experimental results show that the algorithm fully meets the accuracy requirements.In order to solve the problem that the edge of the inner circle of the upper end of the piston is not clear in the process of segmentation,a method of removing the unclear edge by image addition is proposed,and then the diameter of the inner circle of the upper end of the piston is detected by hough circle,and the experiment is carried out with the standard piston to meet the detection accuracy.(5)In the Microsoft Visual Studio 2019 software,the piston size detection system is realized by using C # language programming.The system mainly includes the detection of the piston diameter,the inner edge of the piston pin hole and the inner circle of the upper end of the piston.Each test has a corresponding detection button,and there is an image display area that can show the effect after each step of operation,which can directly display the results after size calculation.