Skeletal Scintigraphy Image Enhancement based Neutrosophic Sets and Salp Swarm Algorithm

Abstract Recently, several schemes are proposed for enhancing the dark regions of the skeletal scintigraphy image. Nevertheless, most of them are flawed by some performance problems. This paper presents an adaptive scheme based on Salp Swarm algorithm (SSA) and a neutrosophic set under multi-criteria to enhance the dark regions of the skeletal scintigraphy image efficiently. Enhancing the dark regions is first converted into an optimization problem. The SSA algorithm is used to find the best improvement for each image separately, and then the neutrosophic algorithm is used to find similarity score to each image with adaptive weight coefficients obtained by the SSA algorithm. The proposed algorithm is applied to an Egyptian medical dataset collected from Menoufia University Hospital and it is a no-reference image. The experiments are done using 3 different resolutions 512*512, 256*256, and 128*128 and compared with Gamma Correction, the NS algorithm and the local enhance algorithm. The results demonstrate that the proposed algorithm achieves superior performance in almost criteria Fitness Function, Entropy, Number of Edges, naturalness image quality evaluator, Sharpness, Sharpness Index, and contrast-distorted images using contrast enhancement. The results showed the idea of integration between the falsity membership of the neutrosophic set and the Salp swarm algorithm can be used to Skeletal Scintigraphy enhancement. This paper proved that it can depend on falsity membership of the neutrosophic set in the Image Enhancement field.