Welcome to the InfoSci Platform
Reference Hub11
Modified Moth-Flame Optimization Algorithm-Based Multilevel Minimum Cross Entropy Thresholding for Image Segmentation

Modified Moth-Flame Optimization Algorithm-Based Multilevel Minimum Cross Entropy Thresholding for Image Segmentation

Abdul Kayom Md Khairuzzaman, Saurabh Chaudhury
Copyright: © 2020 |Volume: 11 |Issue: 4 |Pages: 17
ISSN: 1947-9263|EISSN: 1947-9271|EISBN13: 9781799806721|DOI: 10.4018/IJSIR.2020100106
Cite Article Cite Article

MLA

Khairuzzaman, Abdul Kayom Md, and Saurabh Chaudhury. "Modified Moth-Flame Optimization Algorithm-Based Multilevel Minimum Cross Entropy Thresholding for Image Segmentation." IJSIR vol.11, no.4 2020: pp.123-139. http://doi.org/10.4018/IJSIR.2020100106

APA

Khairuzzaman, A. K. & Chaudhury, S. (2020). Modified Moth-Flame Optimization Algorithm-Based Multilevel Minimum Cross Entropy Thresholding for Image Segmentation. International Journal of Swarm Intelligence Research (IJSIR), 11(4), 123-139. http://doi.org/10.4018/IJSIR.2020100106

Chicago

Khairuzzaman, Abdul Kayom Md, and Saurabh Chaudhury. "Modified Moth-Flame Optimization Algorithm-Based Multilevel Minimum Cross Entropy Thresholding for Image Segmentation," International Journal of Swarm Intelligence Research (IJSIR) 11, no.4: 123-139. http://doi.org/10.4018/IJSIR.2020100106

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Multilevel thresholding is a widely used image segmentation technique. However, multilevel thresholding becomes more and more computationally expensive as the number of thresholds increase. Therefore, it is essential to incorporate some suitable optimization technique to make it practical. In this article, a modification is proposed to the Moth-Flame Optimization (MFO) algorithm and then it is applied to multilevel thresholding for image segmentation. Cross entropy is used as the objective function to select the optimal thresholds. A set of benchmark test images are used to evaluate the proposed technique. The Mean Structural SIMilarity (MSSIM) index is used to measure the quality of the segmented images. The results of the proposed technique are compared with the original MFO, PSO, BFO, and WOA. Experimental results and analysis suggest that the proposed technique outperforms other techniques in terms of segmentation quality images and stability. Moreover, computation time required for multilevel thresholding is also reduced to a manageable level.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.