A novel subpixel edge detection based on the Zernike moment

A novel sub-pixel edge detection and center localization algorithm for micro-parts is presented for improving the resolution and accuracy of micro-assembly systems. Firstly, the invariant moments are employed to descript the feature of the micro parts, then, we extracts the edge of micro parts at pixel level using canny operator. Secondly, Zernike orthogonal moments is used to relocate the edge of micro size part at sub-pixel level. Last, we apply least squares ellipse fitting to position the center. Experiment results show that the algorithm, we presented could achieve higher location accuracy and consume less time. © 2010 Asian Network for Scientific Information.