Abstract:Aiming at the detection problem of aircraft riveting aperture, an improved sub-pixel detection method of riveting aperture based on Zernike moment is proposed. Firstly, the collected riveting aperture image was enhanced by histogram equalization method, smoothed by median filter method and sharpened by gradient method. Then, based on the three image segmentation methods of histogram valley bottom method, Prewitt operator edge extraction and Kirsch operator edge extraction, Canny operator edge detection was used for rough positioning and the improved Zernike moment was used for sub-pixel edge fine extraction to extract the ring area of dimple outer circle. Finally, the circle fitting was carried out in combination with RANSAC algorithm. The circle contour was generated and the radius was obtained to complete the fitting measurement. Experiments show that the detection method has strong sub-pixel accurate positioning ability, high accuracy, good stability and repeated measurement accuracy of ±10 μm.