Abstract: A new measurement method of sprocket end face geometry parameters based on vision principle was proposed aiming at the problems of low efficiency and measurement accuracy in the existing measurement methods of sprocket end face geometry parameters. Firstly,the inner and outer contours of the sprocket were extracted by sub-pixel edge detection of Zernike moments,and Hough transformation was used to obtain the center point coordinates and center hole radius of the inner contour. Secondly,an auxiliary circle was made on the outer contour,and the number of teeth was obtained by the pixel sequence counting algorithm. Then,according to the shape characteristics of the sprocket tooth profile,an improved local radius region algorithm was proposed to extract the tip and root points of each tooth profile,and the least square method was used to fit the pixel points to obtain the tooth apex and tooth root circle diameter. Finally,the least square method was used to fit the sprocket keyway edge line to obtain the keyway size.The repeated measurements were performed 10 times for each of the two sprockets with different specifications. The experimental results show that the measurement errors of sprocket end face geometric parameters are all less than 40 μm,and the relative errors are all within 0.4 %,which can meet the needs of high-precision automatic measurement.

Key words: sprocket, visual measurement, sub-pixel edge, geometric size