Multifractal Morphological Spectra

Transport in Porous Media · Springer Nature · 2026-09-08 · DOI: 10.1007/s11004-026-10334-5

Аннотация

This paper provides a comprehensive approach for the computation of multifractal morphological spectra of spatial functions,$$f\left(x,y\right)$$, spatial sets ($$X)$$, and spatial networks $$S(X)$$ to derive generalized dimensions, including the information dimension. Such spatial functions can depict natural phenomena that are scale-invariant but shape-dependent. Any spatial function $$f\left(x,y\right)$$ can contain rich but concealed information. Such hidden information includes planar forms (spatial sets), ($$X$$) and skeletons (spatial network), $$S\left(X\right)$$. In general, spatial functions $$f\left(x,y\right)$$ are useful for better understanding the spatiotemporal behavior of natural phenomena and their associated features. From a geomorphological perspective, spatial functions,$$f\left(x,y\right)$$, sets, $$X$$, and networks, $$S(X)$$, correspond to digital elevation models (DEMs), threshold elevation regions, and geomorphological networks, respectively. Simulated $$f\left(x,y\right)$$, representing a fractal DEM, produce both the spatial sets, $$X$$, and their skeletons, $$S(X)$$. Mathematical morphological granulometric analysis of the $$f\left(x,y\right)$$ and $$X$$, and the derivation process of the skeleton network $$S(X)$$ generate further useful data, for example, whether the $$f\left(x,y\right)$$, $$X$$ and $$S(X)$$ lose energy in a nonlinear fashion with increasing iterations of multiscale opening with respect to a certain $$B$$ (granulometries), and $$S(X)$$ is derived by uniquely shrinking $$X$$. Such generated data obtained in iterative bisections can be used to compute multifractal morphological spectra that provide different kinds of dimensions, including the information dimension ($${D}_{1}$$), which are $$B$$-specific. This study provides new insights and definitions for the analysis of spatiotemporal complexity.

Читать у издателя →

Похожие статьи