Hoshen–Kopelman algorithm: Difference between revisions

Revision as of 04:30, 11 September 2016

Revision as of 04:41, 11 September 2016

Line 1:

=== Introduction ===

The [[Priestly breastplate|Hoshen]]-Kopelman Algorithm is usedsimple and efficient algorithm for labelinglabelling 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“[[Percolation theory|Percolation]] and cluster distribution. I. Cluster multiple labeling technique and critical concentration algorithm”[1] by J. Hoshen and R. Kopelman.{{dashboard.wikiedu.org sandbox}}

=== 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.{{dashboard.wikiedu.org sandbox}}