A fully parallel iterative thinning algorithm called MB2 is presented. It favourably competes with the best known algorithms regarding homotopy, mediality; thickness, rotation invariance and noise immunity, while featuring a speed improvement by a factor two or more owing to a smaller number of operations to perform. MB2 is grounded on a simple physics-based thinning principle that conveys both quality, efficiency and conceptual clarity. It is particularly suited to data parallel execution.
Thierry M. Bernard, Antoine Manzanera