Study On The Measurement Method Of Part Size Of Automatic Image Measuring Instrument

In this paper,an automatic measurement method of parts size with image measuring instrument is proposed to solve the problems of long detection period and low efficiency of thin plate parts with traditional image measuring instrument.The main research contents of this paper are as follows:In view of the dimension measurement of the parts beyond the field of view,the structure of the traditional image measuring instrument is improved,and the double camera is adopted to solve the problems of the traditional image measuring instrument in the condition of beyond the field of view,such as the large geometric size of the parts to be measured,batch detection and so on.Among them,"big view" camera is used for part identification,and "small view" camera is used for part measurement.In the process of part identification,the part to be tested has arbitrary position and posture relative to the part template.The use of classic matching methods such as grayscale matching or contouring based matching requires the creation of multi-angle template,which results in a large amount of computation and a long matching time.To solve the time-consuming problem of matching,the matching method is improved in this paper.Due to the backlighting method used in large-field imaging,the edges of each part in the image have good contrast,and the shape features of the parts are obvious.Therefore,this paper chooses shape invariant moment feature for part identification.According to the first and second order center distance of the image,the center of gravity difference and principal axis Angle difference between the target part and the template are calculated respectively in this paper.The effectiveness of the method is verified by repeatability experiments.The experimental results show that the binary contour image Hu moment as the feature vector has a good adaptability to the rotation and translation of the measured part in the process of partidentification,with a high accuracy and a short detection time.The detection time of a single image is within 0.05 s.In the dimension measurement of parts,due to the influence of machining accuracy,the edge of parts usually has burr,bump and noise,which affect the measurement accuracy.In order to obtain high precision measurement of parts,the measurement method of parts is studied.By comparing the effect of each sub-pixel edge detection method,the radial basis function interpolation method is selected to extract the sub-pixel edge of the part to be tested to achieve high-precision detection of the geometric features of the part.In the aspect of feature fitting,this paper improves the traditional least squares fitting method,removes the outliers' interference to improve the fitting accuracy of parts' features,and verifies the accuracy of linear and circular feature fitting when there is outliers' interference through experiments.The experimental results show that the radius deviation and center error of the traditional least square fitting circle to be measured are about 0.008 mm and 0.006 mm respectively.The radius deviation of the circle to be measured is about 0.34 microns and the center coordinate deviation is about 0.11 microns.In line feature detection.The standard deviation of the traditional least-squares fitting affected by outliers is 2.6187.After removing outliers,the standard deviation of the fitting deviation is 0.1147,which has higher fitting accuracy.On the basis of this research,the structure of traditional image measurement instrument is improved,a set of principle prototype of automatic image measurement instrument is built,and a set of manual and automatic measurement system for out-of-field of view and multi-size of large quantities of parts is designed.The lastThe method was verified by experiments.The experimental results show that the measurement system is less than 0.005 mm for the dimension measurement accuracy of the parts beyond the field of view.Compared with the traditional image measuring instrument,the automatic image measuring instrument automatically locates the features of the parts,which has a higher measurement efficiency and provides a good solution for the high-precision measurement of the parts beyond the field of view.