Research On Two-dimensional Dimension Measurement Of Motor Parts Based On Machine Vision

With the improvement of science and technology,the manufacturing enterprises continue to improve the production operations,and the requirements for the machining accuracy of the workpiece are continuously improved.Whether the workpiece meets the production requirements cannot be separated from the workpiece dimensional accuracy detection.Machine vision which has the advantages of high precision,fast speed,good flexibility,high level of automation and intelligence is a comprehensive discipline involving many fields such as neurology,biology,psychology,image processing,etc.The online measurement technology based on machine vision which is widely used in medical,biological,industrial,aerospace and other fields is an emerging technology with broad prospects.The research of the workpiece size measurement based on machine vision measurement technology is introduced in this paper and the following works are completed:(1)Visual measurement system is setup and image pre-processing is introduced,then the appropriate hardware equipment and lighting methods is selected.Image preprocessing is performed on the acquired image,including image gray processing,image filtering processing,image contrast stretching and image binarization.By comparing different edge detection operators,Canny with high positioning accuracy and strong noise suppression ability is selected as the operator to extract the edges of the image.(2)In this paper,the feasibility of using neural network in image calibration is analyzed.A BP neural network with double hidden layer is designed.The implicit calibration method is used to image the small field of view measurement environment.The positioning performance and robustness are better.The harris corner detection algorithm extracts the corner points of the checkerboard calibration board,and uses the BP neural network with double hidden layer designed in this paper to complete the image calibration and error analysis.(3)In this paper,the diameter of the motor cover's inner diameter and the length of the standard gauge block are measured by the established visual measurement system,coordinate measuring instrument and vernier caliper.The measurement results show the measurement accuracy ratio based on machine vision measurement technology introduced in this paper can meet the requirements for two-dimensional measurement of motor parts.