In clinical imaging, the fundamental step is to extract useful data from images obtained by different sources and combining to form a single image called fused image to analyse and improve diagnosis. All image fusion techniques now a day rely majorly on non-fuzzy collections. As more qualms are taken into account as compared to non-fuzzy sets, fuzzy sets are better suited for medical image processing. We provide a method for effectively combining multimodal medical images in this study. As suggested, firstly, intuitionistic fuzzy images (IFIs) are obtained considering input source images into account. The optimal membership and non-membership function values are then derived for intuitionistic fuzzy entropy (IFE). After that, the IFIs are scrutinised using the fitness function, contrast visibility (CV). Then, Teaching–Learning-Based Optimization (TLBO) is utilised to maximise the fusion coefficients that are adjusted during the teaching stage and the learning phase of TLBO, allowing the weighted coefficients to naturally adapt to the fitness function. Finally, the fused image is acquired using the optimal coefficients. On various collections of source images, simulations are run, and the output findings are correlated to the most recent fusion techniques. The superiority of the suggested system is explained and supported. Edge-based image fusion (QAB/F), spatial frequency (SF), entropy (E), and other objective metrics are also used to assess how superior the fused image is.
Fuzzy Set-Based Multimodal Medical Image Fusion with Teaching Learning-Based Optimization
Lect. Notes Electrical Eng.
International Conference on Robotics, Control, Automation and Artificial Intelligence ; 2022 November 24, 2022 - November 26, 2022
2023-11-18
16 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Multimodal Medical Image Fusion Using the Sugeno Fuzzy Inference System
Springer Verlag | 2023
|Fuzzy-logic Based Information Fusion for Image Segmentation
British Library Conference Proceedings | 2005
|