Application of evolutionary and swarm optimization in computer vision: a literature survey

Abstract

Evolutionary algorithms (EAs) and swarm algorithms (SAs) have shown their usefulness in solving combinatorial and NP-hard optimization problems in various research fields. However, in the field of computer vision, related surveys have not been updated during the last decade. In this study, inspired by the recent development of deep neural networks in computer vision, which embed large-scale optimization problems, we first describe a literature survey conducted to compensate for the lack of relevant research in this area. Specifically, applications related to the genetic algorithm and differential evolution from EAs, as well as particle swarm optimization and ant colony optimization from SAs and their variants, are mainly considered in this survey.

Classification of Covid-19 chest X-ray images by means of an interpretable evolutionary rule-based approach

“…As far as we know, never have Evolutionary Algorithms (EAs) been used to perform the classification task starting from a data set of images, a fortiori related to Covid-19. This is confirmed by a wide and recent literature survey conducted in 2020 by Nakane et al [ 8 ]. In it, they report that surveys on the application of EAs and swarm algorithms (SAs) to the field of computer vision have not been updated during the last decade, so their paper is the most reliable source on this.…”

Braille Block Detection via Multi-Objective Optimization from an Egocentric Viewpoint

“…To solve multi-objective optimization problems (MOP) with two or three objectives, many multi-objective evolutionary algorithms (MOEA) have been proposed, such as strength Pareto evolutionary Algorithm2 (SPEA2) [ 33 ], NSGA-II [ 2 ], indicator-based evolutionary algorithm (IBEA) [ 34 ], generalized differential evolution 3 (GDE3) [ 35 ], multi-objective evolutionary algorithm with decomposition (MOEA/D) [ 36 ], non-dominated sorting genetic algorithm-III (NSGA-III) [ 37 ], improved decomposition-based evolutionary algorithm (DBEA) [ 38 ], etc. MO algorithms are generally designed to solve problems that require optimizing multiple objectives, and have been applied in the field of computer vision [ 39 ]. For example, Bandyopadhyay et al [ 40 ] proposed land cover classification in remote sensing images with NSGA-II.…”

Coarse-to-Fine Evolutionary Method for Fast Horizon Detection in Maritime Images

“…For quick optimization of the horizon line parameters, GA is used, which provides optimization utilizing fewer combinations of parameters compared to exhaustive search. The GA is broadly applied to efficiently solve combinatorial optimization problems in computer vision such as template matching and object detection [21]- [23].…”

“…Therefore, the problem of the horizon detection can be regarded as a global optimization problem. GA is a popular evolutionary algorithm for global optimization and has been applied to various combinatorial optimization problems in computer vision [21], [23]. Thus, we used the GA to optimize the parameters of horizon line in both coarse and fine optimization.…”