Actively-induced percolation: An effective approach to multiple-object systems characterization

TL;DR

This paper introduces actively-induced percolation as a novel method for characterizing multi-object systems by expanding objects and monitoring cluster formation, effectively capturing spatial interactions.

Contribution

It proposes a new induced percolation approach that uses object expansion and cluster analysis to characterize multi-object systems.

Findings

Effective in synthetic data analysis

Applicable to real-world spatial data

Provides insights into object interactions

Abstract

The present work proposes the concept of induced percolation over multiple-object systems, so that features such as the number of merged clusters can be used as a relevant measurement. The suggested approach involves the expansion of the objects while monitoring the evolving clusters. The potential of the proposed methodology for characterizing the spatial interaction and distribution between several objects is illustrated with respect to synthetic and real data.