STOCHASTIC SIMULATION OF INFRARED NON-GAUSSIAN NATURAL TERRAIN IMAGERY

We describe a new technique for the fast simulation of natural a image ry with non-Gaussian statistics. We pass samples of the imagery throug h an invertible pointwise nonlinear transform, to give samples of a ne w random field with Gaussian marginal probability density. We compute the correlation function for the transformed samples, and simulate Gau ssian random fields (GRFs) with these correlations. Passing the simula ted GRFs through the inverse transform gives our simulations of the or iginal imagery. We discuss the effectiveness of this approach for simu lating natural terrain backgrounds. (C) 1998 Elsevier Science B.V. All rights reserved.