The invention provides a large straight gear geometric parameter measurement
system and method based on
machine vision, and the method comprises the steps: firstly adjusting a camera to a proper position, and enabling the camera to clearly
shoot a needed picture; calibrating the camera to obtain an internal reference matrix and a
distortion vector of the camera; preprocessing the shot tooth profile local image, wherein the preprocessing comprises
distortion correction, image filtering and threshold segmentation; extracting a tooth profile edge local image by using an
edge extraction algorithm, and further finely positioning the edge by using a sub-pixel method to obtain a sub-pixel edge; then, extracting key angular points of the tooth profile by using an
angular point extraction algorithm, and determining a sub-pixel center of the tooth profile according to the angular points; according to the sub-pixel center, obtaining a gear root circle
radius and an addendum circle
radius under the pixel coordinate; and through pixel equivalent calibration, obtaining a real root circle
radius and an addendum circle radius of the gear, thereby calculating a modulus and a reference circle radius of the gear, and completing measurement of geometric parameters of the large straight gear. According to the method, the measurement
automation degree of the geometric parameters of the large straight gear is higher, the function of integrating
monocular camera calibration,
image processing, parameter measurement and calculation can be achieved, the input cost is low, and the
engineering application value is high.