Enhanced multimodal medical image fusion based on Pythagorean fuzzy set- An innovative approach

Abstract

The primary goal of this article is to combine multi-modality medical images into a single output image in order to obtain superior information and better visual appearance without any vagueness and uncertainties, which is suitable for better diagnosis. The complexity of medical images is higher, and many researchers applied various soft computing methods to process them. Pythagorean fuzzy set (PFS) is more suitable for medical images because it considers more uncertainties. In this article, a new method, Pythagorean fuzzy set-based medical image fusion is proposed. Initially, the source images are decomposed into base and detail layers using the two-layer decomposition method, and these layers contain structural and edge details of the source images. To preserve more edge details and clarity, a spatial frequency based fusion rule is employed for detail layers. The base layer images have low contrast, to enhance this; it is converted into Pythagorean fuzzy images (PFIs) with the help of optimum value, which can be generated by Pythagorean fuzzy entropy (PFE). Then, the two pythagorean fuzzy images are decomposed into image blocks, and then perform blackness and whiteness count fusion rule. Finally, the enhanced fused image is obtained by reconstructions of PFI blocks and performs the defuzzification process. The efficiency of the proposed fusion method proves that in terms of both visually and quantitatively compared to other existing fusion methods. The proposed method is tested on different datasets with various quality metrics, which produces an enhanced fused image without artifacts and uncertainties