Open AccessMultilevel Thresholding Image Segmentation based on Autonomous Groups Particle Swarm Optimization
Author(s) -
_ Murinto
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/771/1/012018
Subject(s) - thresholding , particle swarm optimization , image segmentation , artificial intelligence , otsu's method , image (mathematics) , segmentation , pattern recognition (psychology) , multi swarm optimization , computer science , maxima and minima , computation , swarm behaviour , algorithm , mathematics , mathematical optimization , mathematical analysis
Multilevel thresholding problems based on Otsu criteria are discussed in this paper. One weakness of the Otsu method is that computational time increases exponentially according to the number of thresholding dimensions. In this paper, a modified Particle Swarm Optimization (PSO) algorithm called Autonomous Groups Particles Swarm Optimization (AGPSO) is proposed to reduce two problems trapped in local minima and a slow convergence rate in solving high-dimensional problems. AGPSO is used for multilevel thresholding image segmentation. The performance of AGPSO is compared with standard PSO on three natural images. The parameters used to compare the performance of AGPSO and PSO are SSIM, PSNR, Computation Time, optimal threshold obtained from each algorithm. From the experimental results show that AGPSO is better when compared to PSO in image segmentation, from the resulting fitness value and higher SSIM and PNSR values.