Welcome to the InfoSci Platform
Reference Hub26
A New Heuristic Function of Ant Colony System for Retinal Vessel Segmentation

A New Heuristic Function of Ant Colony System for Retinal Vessel Segmentation

Ahmed Hamza Asad, Ahmad Taher Azar, Aboul Ella Hassanien
Copyright: © 2014 |Volume: 1 |Issue: 2 |Pages: 16
ISSN: 2334-4598|EISSN: 2334-4601|EISBN13: 9781466662414|DOI: 10.4018/ijrsda.2014070102
Cite Article Cite Article

MLA

Asad, Ahmed Hamza, et al. "A New Heuristic Function of Ant Colony System for Retinal Vessel Segmentation." IJRSDA vol.1, no.2 2014: pp.15-30. http://doi.org/10.4018/ijrsda.2014070102

APA

Asad, A. H., Azar, A. T., & Hassanien, A. E. (2014). A New Heuristic Function of Ant Colony System for Retinal Vessel Segmentation. International Journal of Rough Sets and Data Analysis (IJRSDA), 1(2), 15-30. http://doi.org/10.4018/ijrsda.2014070102

Chicago

Asad, Ahmed Hamza, Ahmad Taher Azar, and Aboul Ella Hassanien. "A New Heuristic Function of Ant Colony System for Retinal Vessel Segmentation," International Journal of Rough Sets and Data Analysis (IJRSDA) 1, no.2: 15-30. http://doi.org/10.4018/ijrsda.2014070102

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The automatic segmentation of blood vessels in retinal images is the crucial stage in any retina diagnosis systems. This article discussed the impact of two improvements to the previous baseline approach for automatic segmentation of retinal blood vessels based on the ant colony system. The first improvement is in features where the length of previous features vector used in segmentation is reduced to the half since four less significant features are replaced by a new more significant feature when applying the correlation-based feature selection heuristic. The second improvement is in ant colony system where a new probability-based heuristic function is applied instead of the previous Euclidean distance based heuristic function. Experimental results showed the improved approach gives better performance than baseline approach when it is tested on DRIVE database of retinal images. Also, the statistical analysis demonstrated that was no statistically significant difference between the baseline and improved approaches in the sensitivity (0.7388± 0.0511 vs. 0.7501±0.0385, respectively; P = 0.4335). On the other hand, statistically significant improvements were found between the baseline and improved approaches for specificity and accuracy (P = 0.0024 and 0.0053, respectively). It was noted that the improved approach showed an increase of 1.1% in the accuracy after applying the new probability-based heuristic function.

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.