Colour Quantisation using Human Mental Search and Local Refinement

摘要: Colour quantisation is a common image processing technique to reduce the number of distinct colours in an which are then represented by colour palette. Selection appropriate entries this palette challenging since quality quantised directly dictated colours. In paper, we propose novel algorithm based on human mental search (HMS) and subsequent refinement using k-means. HMS recent population-based metaheuristic that has been shown yield good performance variety optimisation problems. first stage, use find high-quality initial second refined k-means converge towards local optimum thus further improve image. We evaluate our set benchmark images compare it several conventional soft computing-based algorithms demonstrate excellent quality, outperforming other methods.