Design Of Dimension Measurement System For Cylindrical Parts Based On Machine Vision

With the development of intelligent manufacturing technology,people are yearning for intelligent automatic size measurement.The traditional dimension measurement mostly uses the measuring tool to measure the parts directly,the measurement process is time-consuming and laborious,easy to damage the workpiece,with the development of machine vision measurement technology,new measurement methods are more and more popular.Cylindrical parts have a wide range of applications in the machinery industry,its size value qualified or not will directly affect the stability and life of parts and equipment.At present,the dimension measurement of cylindrical parts is mostly manual single dimension measurement,which is difficult to meet the requirements of rapid and accurate dimension measurement in intelligent automation production.In view of the above problems,this paper takes the cylindrical parts as the object,studies and designs a set of cylindrical parts size measurement system based on machine vision.Using this system,the height and diameter of several cylindrical parts can be measured synchronously,which overcomes the shortcomings of traditional manual single size measurement.The main research contents of this paper are as follows:(1)Design a set of cylindrical parts size measurement system.According to the characteristics of cylindrical parts size measurement,the hardware part selection and design and software part design are completed.The selection and design of the hardware part mainly includes the selection of the industrial camera,lens,light source,stepper motor and the design of the rotating disc and chassis.The design of the software part includes the realization of the image processing algorithm and the design of the human-computer interaction interface.(2)The algorithm of image preprocessing and the algorithm of obtaining the minimum enclost rectangle of image are studied.Of image filtering,image enhancement,binarization threshold segmentation,edge detection algorithms are studied,through the different image preprocessing algorithm for comparison and analysis of experimental result,choose the most suitable for the algorithm of this system and the corresponding parameters,including selecting median filter for image denoising,using gray linear transformation improve image contrast,choose the binarization threshold segmentation implementation,using Canny operator for edge detection,image segmentation using Andrew convex hull scanning method and rotating block method for cylindrical parts image of the minimum circumscribed rectangle,etc.,Finally,the dimensions of cylindrical parts are measured by measuring the minimum outer rectangle of the image.(3)The distortion of industrial camera is corrected,and the method of error compensation is proposed to compensate the pixel equivalent of the measurement system.By using the internal parameters and distortion coefficient of the industrial camera to correct the image distortion,the error compensation value of the pixel equivalent of the measurement system was obtained through the repeatability measurement of the standard gauge block,which was used to compensate the pixel equivalent of the measurement system.(4)A measurement platform for cylindrical parts was built,and a common cylindrical bearing roller was taken as the measurement object to realize the synchronous measurement of the height and diameter of multiple cylindrical rollers.Halcon12 combined with C# programming is used to complete the design of human-computer interaction interface.The experimental results show that the average relative errors between the height and diameter of the cylindrical roller are small,which realizes the dimension measurement of cylindrical parts based on machine vision.