Multi-size Precision Measurement System Of Electric Connector Shell

High precision inspection of parts is an important part of ensuring the quality of parts. Nowadays, most manufacturing companies still use artificial methods or simple tools to measure, which is slow and difficult to ensure accuracy. Machine vision has the advantages of non-contact, high precision and rapidity,it has become an important development direction and research focus of precision measurement.This paper makes the electric connector shell parts measurements of enterprises production as the research and studies on such aspects as camera calibration, image processing and edge detection, and finally designs a multidimension precision measuring system for electric connector shell parts based on machine vision.Measurement accuracy can meet the actual requirements of enterprises.The main research contents of this paper are as follows:(1)On the base of deeply understanding of vision measurement theory and the research status at home and abroad, this paper has finishedthe hardware selection of the measurement system according to the actual demand and set up a visual measurement platform including CMOS camera, far heart lens, light source andobjective table.(2)Starting from the four coordinate system, studies the camera model and the existing camera calibration method, this paper usesa fast camera calibration method based on HALCON. The experiment shows that the method has high calibration precision, accurate calibration resultsand high computing efficiency.(3)This paper studies the image processing technology such as image binaryzation, image filtering, image segmentation and edge detection, analysesthe basic principles of various methods and provides various of the image processing effects of parts. Through the experimental analysis, the image processing algorithm of the measurement system is established, and it is ready for the following measurement work. This paper has conducted the research focusing on the sub-pixel edge detection technology and use a sub-pixel edge detection method based on the two degree polynomial interpolation to realize the sub pixel edge localization.Experiments show that the method can reduce the measurement error and improve the repeatability measurement accuracy of the system. An improved Harris corner detection method is proposed to realize the sub pixel level localization of corner points and improve the measurement accuracy of the length of the line. The measurement of the diameter and pitch of the holes is realized by using the least squares fitting method and obtaining the circle center coordinates.The measurement of angle parameters is realized by using the angle between straight lines.(4)The functional software and human computer interface of measurement system is designed based on Visual C++6.0 and HALCON, it can realize the functions of image acquisition, image processing, dimensional measurement,exporting results and so on.(5)Measurements are made for three typical electrical connector housing parts. The experimental data analysis that the measurement accuracy of the measurement system can meet the requirements of the enterprise 0.05 mm.