Open AccessModified salp swarm algorithm based multilevel thresholding for color image segmentation
Author(s) -
Shikai Wang,
Heming Jia,
Xiaoxu Peng
Publication year - 2020
Publication title -
mathematical biosciences and engineering
Language(s) - Uncategorized
Resource type - Journals
SCImago Journal Rank - 0.451
H-Index - 45
eISSN - 1551-0018
pISSN - 1547-1063
DOI - 10.3934/mbe.2020036
Subject(s) - image segmentation , fitness function , artificial intelligence , pattern recognition (psychology) , swarm behaviour , thresholding , segmentation , entropy (arrow of time) , algorithm , feature (linguistics) , mathematics , computer science , image (mathematics) , mathematical optimization , genetic algorithm , linguistics , philosophy , physics , quantum mechanics
This paper proposes a multi-threshold image segmentation method based on modified salp swarm algorithm (SSA). Multi-threshold image segmentation method has good segmentation effect, but the segmentation precision will be affected with the increase of threshold number. To avoid the above problem, the slap swarm optimization algorithm (SSA) is presented to choose the optimal parameters of the fitting function and we use levy flight to improve the SSA. The solutions are assessed using the Kapur's entropy, Otsu and Renyi entropy fitness function during the optimization operation. The performance of the proposed algorithm is evaluated with several reference images and compared with different group algorithms. The results have been analyzed based on the best fitness values, peak signal to noise ratio (PSNR), and feature similarity index measures (FSIM). The experimental results show that the proposed algorithm outperformed other swarm algorithms.