Edge Detection Method for Medical Imaging Using Discrimination Measure and Entropy on Pythagorean Fuzzy Environment

In this paper, we develop an edge detection model based on Pythagorean fuzzy sets (PFSs) for dealing with uncertainty in images before edge detection. In this regard, novel Pythagorean fuzzy (PF)-entropy and discrimination measures are developed to enumerate the degree of difference between objects or images, which can evade the limitations of existing PF-discrimination measures. Moreover, several elegant properties of the developed PF-discrimination and entropy measures are presented in the context of PFSs. Numerical examples are presented to illustrate the efficacy of the developed PF-entropy and discrimination measures over the existing PF-discrimination measures. Further, an edge detection model is presented with the developed PF-discrimination measures and entropy and its application in edge detection. To show the usefulness of the developed edge detection approach on PFSs, the computational outcomes are compared with existing edge detection models. Finally, some measurement parameters such as mean square error (MSE), peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) are determined. The findings of this study conclude that the PSNR and SSIM values of the developed model are continuously greater than or equal to the PSNR and SSIM values of various extant edge detection models, respectively. The obtained results on the medical images have shown their applicability to detect edges appropriately in the occurrence of uncertainty and in the presence of noise. Experimental results on several images validate the theoretical results and demonstrate good performance of the developed model.