The Roundness Measurement of Small Parts Based on Machine Vision
Aiming at the problems such as low efficiency and unstable accuracy of roundness error meas-urement for small parts,a method of roundness measurement for small parts based on machine vision was proposed.Through the establishment of machine vision inspection platform,the parts image is obtained.Af-ter pre-processing,the region of interest is delimited by Canny operator combined with 8-domain expansion algorithm,and the edge coordinates are obtained by polynomial interpolation subpixel edge detection algo-rithm.Then,the improved region search algorithm is used to determine the quasi-center of the circle,and the minimum center of the circle is found by the geometric structure of the minimum region method,and the roundness error is calculated.The experimental results show that the repeatability of roundness error is 8. 1 μm,which can realize the non-contact measurement of small parts.
machine visionroundness erroredge detectionminimum zone circlesmall parts
NETL
NSTL
万方数据
维普
