Authors :
Saswati Sahoo; Suman Bala Behera
Volume/Issue :
Volume 7 - 2022, Issue 7 - July
Google Scholar :
https://bit.ly/3IIfn9N
Scribd :
https://bit.ly/3Q5sW7C
DOI :
https://doi.org/10.5281/zenodo.6965411
Abstract :
In image segmentation field Multilevel
thresholding is an important technique. However, in
standard methods, the complexity of this method
increases with the variation of number of thresholds
value. To avoid this disadvantages, nature inspired
meta-heuristic techniques are used. These metaheuristic algorithms give near exact results in a
reasonable time, which catches the attention of recent
researchers for optimization. No matter what kind of
optimization method , the solution set must be
represented via some way. For example, GWO (Grey
Wolf Optimizer) this method follows the grouping and
hunting behavior of wolves, PSO (Particle Swarm
Optimizer) inspired from foraging behavior of swarm of
particles (assuming birds as particles). Above optimizer
are applied on some standard images collected from
USC SIPI, BSD 500 database. At end part, a
comparison was made based on threshold values, image
quality measures, and computational time .In this
analysis, GWO based results and PSO based results are
compared with the standard results .The results are
compared in terms of thresholded images, image quality
.Time complexity and plotting convergence plots, which
shows the goodness of Grey Wolf Optimizer in terms of
global search.
Keywords :
Segmentation, thresholding, particle swarm, meta-heuristic, optimization
In image segmentation field Multilevel
thresholding is an important technique. However, in
standard methods, the complexity of this method
increases with the variation of number of thresholds
value. To avoid this disadvantages, nature inspired
meta-heuristic techniques are used. These metaheuristic algorithms give near exact results in a
reasonable time, which catches the attention of recent
researchers for optimization. No matter what kind of
optimization method , the solution set must be
represented via some way. For example, GWO (Grey
Wolf Optimizer) this method follows the grouping and
hunting behavior of wolves, PSO (Particle Swarm
Optimizer) inspired from foraging behavior of swarm of
particles (assuming birds as particles). Above optimizer
are applied on some standard images collected from
USC SIPI, BSD 500 database. At end part, a
comparison was made based on threshold values, image
quality measures, and computational time .In this
analysis, GWO based results and PSO based results are
compared with the standard results .The results are
compared in terms of thresholded images, image quality
.Time complexity and plotting convergence plots, which
shows the goodness of Grey Wolf Optimizer in terms of
global search.
Keywords :
Segmentation, thresholding, particle swarm, meta-heuristic, optimization
