An Improved Firefly Algorithm-Based 2-D Image Thresholding for Brain Image Fusion

Srikanth M. V., V. V. K. D. V. Prasad, K. Satya Prasad

Source Title: Research Anthology on Improving Medical Imaging Techniques for Analysis and Intervention

ISBN13: 9781668475447|ISBN10: 1668475448|EISBN13: 9781668475454

DOI: 10.4018/978-1-6684-7544-7.ch018

MLA

Srikanth M. V., et al. "An Improved Firefly Algorithm-Based 2-D Image Thresholding for Brain Image Fusion." Research Anthology on Improving Medical Imaging Techniques for Analysis and Intervention, edited by Information Resources Management Association, IGI Global, 2023, pp. 341-379. https://doi.org/10.4018/978-1-6684-7544-7.ch018

APA

Srikanth M. V., Prasad, V. V., & Prasad, K. S. (2023). An Improved Firefly Algorithm-Based 2-D Image Thresholding for Brain Image Fusion. In I. Management Association (Ed.), Research Anthology on Improving Medical Imaging Techniques for Analysis and Intervention (pp. 341-379). IGI Global. https://doi.org/10.4018/978-1-6684-7544-7.ch018

Chicago

Srikanth M. V., V. V. K. D. V. Prasad, and K. Satya Prasad. "An Improved Firefly Algorithm-Based 2-D Image Thresholding for Brain Image Fusion." In Research Anthology on Improving Medical Imaging Techniques for Analysis and Intervention, edited by Information Resources Management Association, 341-379. Hershey, PA: IGI Global, 2023. https://doi.org/10.4018/978-1-6684-7544-7.ch018

Export Reference

Favorite

View Full Text HTML

View Full Text PDF

Abstract

In this article, an attempt is made to diagnose brain diseases like neoplastic, cerebrovascular, Alzheimer's, and sarcomas by the effective fusion of two images. The two images are fused in three steps. Step 1. Segmentation: The images are segmented on the basis of optimal thresholding, the thresholds are optimized with an improved firefly algorithm (pFA) by assuming Renyi entropy as an objective function. Earlier, image thresholding was performed with a 1-D histogram, but it has been recently observed that a 2-D histogram-based thresholding is better. Step 2: the segmented features are extracted with the scale invariant feature transform (SIFT) algorithm. The SIFT algorithm is good in extracting the features even after image rotation and scaling. Step 3: The fusion rules are made on the basis of an interval type-2 fuzzy set (IT2FL), where uncertainty effects are minimized unlike type-1. The novelty of the proposed work is tested on different benchmark image fusion data sets and has proven better in all measuring parameters.