abstract

• Honeybees, as social insects, follow a modular strategy applied to dynamic environments to provide reasonable opportunities for partial solutions to evolve in the form of interacting coadapted subcomponents. The honeybee search algorithm combines concepts from the areas of evolutionary algorithms and swarm intelligence to solve optimization problems. This algorithm is mainly based on the foraging behavior of honeybees and the search power of evolution strategies, a type of evolutionary algorithm used for real-valued problems. This paper shows the integration between an automaton and the honeybee search algorithm to formalize the algorithm mathematically. The combination mentioned above is tested here with the innovative applications of three-dimensional scene reconstruction and video tracking. The experimental results for both applications show evidence that the honeybee search algorithm can be used to improve time costs in challenging computer vision tasks through controlled experiments and objective comparisons. Also, the validation of results demonstrates that the measured accuracy ranks top-tier among other algorithms in the ALOV benchmark. © 2020
• Honeybees, as social insects, follow a modular strategy applied to dynamic environments to provide reasonable opportunities for partial solutions to evolve in the form of interacting coadapted subcomponents. The honeybee search algorithm combines concepts from the areas of evolutionary algorithms and swarm intelligence to solve optimization problems. This algorithm is mainly based on the foraging behavior of honeybees and the search power of evolution strategies, a type of evolutionary algorithm used for real-valued problems. This paper shows the integration between an automaton and the honeybee search algorithm to formalize the algorithm mathematically. The combination mentioned above is tested here with the innovative applications of three-dimensional scene reconstruction and video tracking. The experimental results for both applications show evidence that the honeybee search algorithm can be used to improve time costs in challenging computer vision tasks through controlled experiments and objective comparisons. Also, the validation of results demonstrates that the measured accuracy ranks top-tier among other algorithms in the ALOV%2b%2b benchmark. © 2020

authors

• Castillo Barrera Francisco Edgar
• Núñez Varela José Ignacio
• Olague G.
• Perez-Cham O.E.
• Puente Montejano César Augusto
• Soubervielle Montalvo Carlos

publication date

• 2021-01-01

funding provided via

• Centro de Investigación Científica y de Educación Superior de Ensenada, Baja California, CICESE: 634-135; Consejo Nacional de Ciencia y Tecnología, CONACYT: 25096, CB-2012/177041  Grant

published in

• Swarm and Evolutionary Computation  Journal

keywords

• Evolutionary algorithms; Meta-heuristic; Swarm intelligence; Three-dimensional scene reconstruction; Video tracking Evolutionary algorithms; Image reconstruction; Learning algorithms; Robots; Video recording; 3D scene reconstruction; Controlled experiment; Dynamic environments; Evolution strategies; Foraging behaviors; Optimization problems; Search Algorithms; Three-dimensional scenes; Three dimensional computer graphics

Digital Object Identifier (DOI)

• 10.1016/j.swevo.2020.100817

PubMed ID

start page

end page

volume

• 61

issue