A new method for image processing using generalized linguistic neutrosophic cubic aggregation operator

Abstract

Image processing is a technique of scrutinizing an image and extricating important information. Indecisive situations are generally undergone when the picture processes with profuse noise. Neutrosophic set (NS), a part of neutrosophy theory, studies the scope of neutralities and is essential to reasoning with incomplete and uncertain information. However, the linguistic neutrosophic cubic set (LNCS) is one of the extensions of the NS. In LNCS, each element is characterized by the interval-valued and single-valued neutrosophic numbers to handle the data uncertainties. Keeping these features in mind, we apply LNCS for image processing after defining their aggregation operators and operations. In this study, noisy grey-scale images were transformed into the LNCS domain using three membership degrees, then aggregated using aggregation operators. The proposed method clarifies the noise in the Lena image and three other test images. It has justified the utilization of operators based on visual clarity obtained. Suitable comparison analysis and efficiency testing is performed on the proposed theory by considering noise types, such as Gaussian, Poisson, and Speckle. In addition, we have also compared the computational efficiency of our proposed method with existing ones. The results show that our approach consumes less memory and executes quicker than the existing methods. A decision-maker can select a more effective operator to segment the images more effectively using the obtained results.