Edge detection of screw thread based on machine vision

Abstract

Aiming at the problem of thread measurement with small pitch, a screw thread parameters measurement method based on machine vision is proposed. Which solves the problem that the tip of the mechanical scanning probe is too large to measure the small size pitch diameter. This work developed a method of image processing method to detect the edge of screw thread parameters. The screw thread image is collected by calibrating the camera to correct the distortion error. Canny edge detection is used to detect the edge contours in the thread image, And mean filtering and median filtering. Gaussian smoothing filtering is also studied to obtain higher measurement accuracy; a subpixel subdivision technique is designed to improve the resolution of the imaging system. In order to rotate the screw thread image, a minimum rectangle fitting algorithm for the measurement area is developed. The affine transformation matrix is constructed by the center of the measurement area and the orientation of the long axis of the screw thread. This thread parameter edge detection technique has the advantages of simple and quick operation and high measurement accuracy.