Research On Averaging Glass Volume Automatic Cutting Technology Based On Machine Vision

The optical element is molded from a block blank,which is cut from optical glass strips according to determined weight.The traditional cutting process has poor accuracy,slow speed,and low degree of automation and can no longer meet actual needs.In recent years,machine vision inspection technology has been widely used in various industrial processes due to its high accuracy,fast speed,and non-contact characteristics.The thesis designs and implements a machine vision inspection for a company’s existing optical glass weight-based automatic cutting system.Thesis mainly completes the following work;1.A simple method based on plane mapping matrix H is used to calibrate the visual system of the loading manipulator.This method is based on the idea of mapping the image coordinates to the coordinates of the manipulator,and does not solve the camera parameter values,simplifying the calibration process.Experiments show that the calibration method is easy to operate,takes 38 seconds,and the accuracy is of 0.01 magnitude.2.A corner detection method based on Gabor wavelet transform is used to calibrate the relationship between the image pixel area and the actual area,The Gabor wavelet transform has excellent characteristics in describing image texture features and it improves the accuracy of corner detection.3.The glass average weight cutting algorithm vision-based is designed and software implementation based on OpenCV library functions is completed.Bilateral filtering and canny operator suitable for this thesis are selected.Calibrate the relationship between the image pixel area and the area of the glass strip end surface area and output the cut length data.Experimental results show that the glass block weight error is controlled within ±3% of the allowable error range,and the pass rate is above 93%.