Massive Conscious Neighborhood-Based Crow Search Algorithm for the Pseudo-Coloring Problem

Abstract

The pseudo-coloring problem (PsCP) is a combinatorial optimization challenge that involves assigning colors to elements in a way that meets specific criteria, often related to minimizing conflicts or maximizing some form of utility. A variety of metaheuristic algorithms have been developed to solve PsCP efficiently. However, these algorithms sometimes struggle with the quality of solutions, impacting their ability to achieve optimal or near-optimal results reliably. To overcome these issues, this study introduces an adapted conscious neighborhood-based crow search algorithm (CCSA) and a massive variant of CCSA specifically tailored for PsCP. The performance of CCSA and MCCSA are evaluated on real and synthetic images and compared with state-of-the-art optimizers. The results showed that the adapted CCSA and MCCSA outperformed offering an effective strategy for image pseudo-colorization.