Variable neighbourhood search for colour image quantization

Abstract

Colour image quantization is a data compression technique that reduces the total set of colours in a digital image to a representative subset. This problem is first expressed as a large M-median one. The advantages of this model over the usual minimum sum-of-squares model are discussed first and then, the heuristic based on variable neighbourhood search metaheuristic is applied to solve it. Computational experience proves that this approach compares favourably with two other recent state-of-the-art heuristics, based on genetic and particle swarm searches.