Autofocusing algorithm of interferogram based on object image and registration technology

Because the depth of field (DOF) of lenses is limited, a captured interferogram usually includes a focused region and defocused region simultaneously when the form deviation of the helical tooth flank is measured with laser interferometry. Because of the impact of the interference fringe, the existing autofocusing algorithms are difficult for identifying the focused region of the interferogram directly. This paper proposes a new autofocusing algorithm based on the object image and registration technology. First, the reliability region is evaluated according to the gray level. Then the corresponding pixels in the image sequence are determined with the registration technology. Next, the focused region is judged by comparing the average gray gradients in object images. Based on the characteristics of measured gear and the DOF of lens, an experiment is designed. The experimental results are given to verify the feasibility and accuracy of the proposed autofocusing algorithm.