Image Segmentation Using Kapur’s Entropy and a Hybrid Optimization Algorithm

Diego Oliva, Mohamed Abd Elaziz, Salvador Hinojosa

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

Image thresholding has different limitations especially with the methods used to search the best configuration of thresholds. To avoid these drawbacks, the meta-heuristic algorithms are commonly used, they have the ability to find the global solution in a reduced number of iterations. Based on this concept, this chapter presents an improvement of the salp swarm algorithm based on artificial bee colony as an alternative image segmentation method. The proposed method combines the operators of the ABC with SSA and this lead to improve the convergence and find the best threshold value. The proposed method, called SSAABC, uses the Kapur’s function to assess the quality of each solution. In order to evaluate the performance of the proposed approach six images are used as test and the results are compared with four different algorithms. The Experimental results provides an evident about the high performance of the proposed SSAABC method in terms of the performance measures such as PSNR, SSIM, and CPU time(s).

Original languageEnglish
Title of host publicationStudies in Computational Intelligence
PublisherSpringer Verlag
Pages85-99
Number of pages15
DOIs
Publication statusPublished - 2019
Externally publishedYes

Publication series

NameStudies in Computational Intelligence
Volume825
ISSN (Print)1860-949X

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Image Segmentation Using Kapur’s Entropy and a Hybrid Optimization Algorithm'. Together they form a unique fingerprint.

  • Cite this

    Oliva, D., Abd Elaziz, M., & Hinojosa, S. (2019). Image Segmentation Using Kapur’s Entropy and a Hybrid Optimization Algorithm. In Studies in Computational Intelligence (pp. 85-99). (Studies in Computational Intelligence; Vol. 825). Springer Verlag. https://doi.org/10.1007/978-3-030-12931-6_8