QUANTIFYING MORPHOLOGY OF SCALE-INVARIANT STRUCTURES BEYOND THE FRACTAL DIMENSION
We present a correlation scheme to quantify the morphology beyond the standard fractal dimension which corresponds to information from the pair correlation function. The method consists of analyzing hierarchical correlations in log-space thus summing contributions from higher order correlations in the usual space coordinates. The scheme gives information on the characteristics of structure which can be used as a fingerprint to distinguish between structures with the same fractal dimension. The method is also sensitive to oscillations in logarithmic scales, which are admissible solutions for renormalisation equations. Such oscillations appear as resonances, thus making this scheme suitable to analyze such phenomena experimentally. Illustrative examples are given for all those applications by analyzing numerically grown structures. The case of the unrestricted Brownian walk is exactly calculated. We discuss an application of this scheme to check recent analytic results obtained for scale-invariant branching mechanisms in slow cracking patterns and in noise-reduced diffusion-limited-aggregation. We propose that this method is a suitable candidate to quantify the presently qualitative concept of morphology.