Method Of Measuring Male Thread Parameters Based On Image Detection Technology And System Development

The use of externally threaded parts,both in daily life and in industrial production,is extensive,and usually it is used in conjunction with internal thread components to connect,transmit,and tighten.The external thread parts are inexpensive and have a wide range of applications.Therefore,the reliability of external thread parts is very important for the normal operation of mechanical equipment.Experts and scholars have conducted in-depth studies on the measurement methods and methods of external threads.Under the conditions of sophisticated testing equipment,the speed and accuracy of external thread industrial production can reach relatively high requirements.The rapid and accurate measurement of the geometric parameters also puts forward an urgent need.Experts and scholars both at home and abroad have conducted extensive research on detecting threads using image detection techniques for external thread measurement methods,slow thread detection speeds,and low accuracy,but the preconditions for satisfying demand are the use of expensive testing equipment.This project selects ordinary bolts as the research object,uses a relatively low cost to build an economical and practical inspection platform,and aims to achieve non-contact,high-efficiency,high-precision external thread detection as the research goal,and proposes a method based on image detection technology to measure Thread parameter method,and analyze the experimental results.Among them,the research is mainly based on the design,detection algorithm and system error of the image detection system scheme.Firstly,the camera is used to capture the four-image image acquisition scheme during the rotation of the camera with the fixed thread rotation.By comparing the four faces of the thread,the problem of insufficient accuracy due to the detection of only one face of the externally threaded part is avoided.Then the related algorithms of image processing and geometric parameters extraction of external thread are studied.First,the thread is preprocessed.Based on the preprocessing,the image is thresholded according to the gray histogram of the image,and then the external thread image is used.The edge characteristics of the edge selection of the appropriate edge detection operator,and proposed the use of least squares method to fit the regression line to fit the middle axis of the thread,the thread from the outer thread edge of the line to the outer thread axis to calculate the distance The four major parameters of the major diameter,middle diameter,minor diameter,and pitch,and use these four external thread parameters as the target of visual inspection of thread parameters.Finally,the summary of the experimental data and the analysis of the error are performed.It can be concluded from the error analysis results that under the condition of limited experimental conditions and inaccurate mechanical installation,the measurement error is also smaller than the national standard tolerance.Therefore,the method proposed in this paper can realize the non-contact thread geometry detection and the detection accuracy.There has also been a certain degree of improvement.