Narrow-Band Mask Synthesis with Semi-Implicit Difference

Abstract

In this paper, a distance level-set regularized reformulation of mask synthesis is developed to secure a simple and straightforward construction of the narrow band and provide the nonlinear diffusion term for implicit difference schemes. Subsequently, the mask update is performed only in the vicinity of the zero level set thereby reducing optimization dimensionality; moreover, the semi-implicit discretization is applied to circumvent the stability constraints enabling sufficiently large stepsize improving convergence with much less iteration numbers. Additive operator splits the mask update with respect to coordinate axes to solving multiple comparatively small scale linear systems of equations with Thomas method. Simulation results merit the superiority of the proposed approach with improved convergence by overcoming the stability constraints and reduced optimization dimensionality.