An algorithm of subpixel edge detection based on ZOM and application in calibration for robot vision

An algorithm of subpixel edge detection based on a Zernike orthogonal moment (ZOM) is presented. The ZOM not only may detect the line edge but also may detect the curve edge. In the paper the properties of the ZOM are analyzed and a gray step model of an image is established taking a line edge as an example. The computing formulas for subpixel edges are gained after analyzing the principle of subpixel edge detection. Three orthogonal moments only are required for subpixel edge detection. After computing the amplitude and vertical direction of the edge, the maximum of the amplitude is searched along the vertical direction in order to locate the subpixel edge. A detailed algorithm is provided. Experimental results show that subpixel edge detection on ZOM has high locating precision and strong robustness to noise.