Hoshen–Kopelman algorithm

This is an old revision of this page, as edited by Dssathe (talk | contribs) at 04:41, 11 September 2016 (percolation theory). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Introduction

The Hoshen-Kopelman Algorithm is simple and efficient algorithm for labelling clusters on a grid, where grid is a regular network of cells, where each cell may be “occupied” or “unoccupied”. This algorithm is based on well-known union-finding algorithm. The algorithm was originally described in “Percolation and cluster distribution. I. Cluster multiple labeling technique and critical concentration algorithm”[1] by J. Hoshen and R. Kopelman.

Percolation Theory

Percolation theory is the study of the behavior and statistics of clusters on lattices. Suppose we have a large square lattice where each cell can be occupied with the probability p and empty with probability 1 – p.

Each group of neighboring occupied cells forms a cluster. Neighbors are defined as cells having a common side but not those sharing only a corner i.e. we consider 4x4 neighborhood. (top, bottom, left, right). Each occupied cell is occupied independently of the status of its neighborhood. The number of clusters, size of each cluster and their distribution are important topics in percolation theory.

This template should only be used in the user namespace.This template should only be used in the user namespace.