Improving a method for segmenting optical-electronic images acquired from space observation systems based on the firefly algorithm

Abstract

The object of research is the process of segmentation of optoelectronic images acquired from space observation systems. The method of segmentation of optoelectronic images acquired from space observation systems based on the firefly algorithm, unlike known ones, involves the following: – the pre-selection of brightness channels of the Red-Green-Blue color space in the original image; – calculation of the level of luminosity for each firefly; – assigning each firefly with the neighboring firefly, within a certain radius, whose level of luminosity is higher than the natural level of luminosity of the firefly; – determination of the coordinate of the updated position of the firefly in each brightness channel. An experimental study into the segmentation of optoelectronic images acquired from space observation systems based on the firefly algorithm was carried out. It is established that the improved segmentation method based on the firefly algorithm allows for the segmentation of optoelectronic images acquired from space observation systems. The quality of segmentation of optoelectronic images by the method based on the firefly algorithm was evaluated in comparison with methods based on the particle swarm algorithm and the Sine-Cosine algorithm. It was found that the improved method based on the firefly algorithm reduces the segmentation error of the first kind by an average of 11 % and the segmentation error of the second kind by an average of 9 %. This becomes possible by using the firefly algorithm. Methods of image segmentation can be implemented in software and hardware systems for processing optoelectronic images acquired from space surveillance systems. Further studies may focus on comparing the quality of segmentation method based on the firefly algorithm with segmentation methods based on genetic algorithms.