Estimating the fractal dimension of synthetic topographic surfaces

摘要

This paper reports an analysis comparing several statistical methods for the estimation of the fractal dimension D of computer generated synthetic surfaces. Ideally, such methods must be robust and accurate. This cannot be determined without direct comparison of the performance of both (a) different methods and (b) different estimation algorithms for each method. Results are presented for the analysis of simulated data using three estimators of the 1D semivariance function, and the 1D maximum entropy power spectral density to derive D. These suggest that method-produced variation can limit the utility of D as an index of discrimination. (C) 1998 Elsevier Science Ltd. All rights reserved. References: 39