Fast artificial bee colony and its application to stereo correspondence

摘要: A novel optimisation method.Apply to stereo matching.Outperform the current methods. Nature-inspired meta-heuristics have gained popularity for solutions many real-world complex problems, and artificial bee colony algorithm is one of most powerful methods among meta-heuristics. However, inefficient exploitation onlooker bees prevents from finding final result accurately efficiently problems. In this paper, a method proposed based on algorithm. The adaptively exploits over generations. addition, applied stereo-matching problem minimise segment-based integer energy function, which also introduced in paper. experimental results show that outperforms state-of-the-art population-based meta-heuristics, such as genetic algorithm, differential evolution, conventional colony, clonal selection benchmark functions well problem.