A robust edge-based template matching algorithm for displacement measurement of compliant mechanisms under scanning electron microscope.

Abstract

This paper develops a robust edge-based template matching algorithm for displacement measurement of compliant mechanisms under a scanning electron microscope (SEM). The algorithm consists of three steps. First, the Sobel gradient operator and a self-adaptive segment strategy are used to establish the shape model in which the gradient directions of the object's edge points are calculated. Second, a similarity criterion based on image gradients that is robust to illumination change and image noise is utilized for template matching to obtain the coarse results. The third step is to refine the matching results by using an orientation-guided subpixel interpolation strategy. A series of simulations is conducted, and the results show that the proposed algorithm enjoys great robustness against strong image noise and gray-value fluctuation, as well as small rotations and background interferences, and thus is suitable for processing SEM images of compliant mechanisms. Finally, the application of the proposed algorithm in the measurement of the spring constant of the flexure hinges with a straight beam form under a SEM is demonstrated.