Jaya based functional link multilayer perceptron adaptive filter for Poisson noise suppression from X-ray images

In this paper, a parameterless Jaya optimization based neural network filter named as Jaya-functional link multilayer perceptron (Jaya-FLMLP) is proposed for the elimination of Poisson noise from X-ray images. In this proposed adaptive filter, Jaya is applied for updating the weights of the FLMLP network. The proposed neural filter is a combination of a functional link artificial neural network (FLANN) and Multilayer Perceptron (MLP) network. The performance of Jaya-FLMLP is also compared with other five competitive networks such as Wiener, MLP, Least Mean Squares based Functional Link Artificial Neural Network (LMS-FLANN), Particle Swarm Optimization based Functional Link Artificial Neural Network (PSO-FLANN) and Cat Swarm Optimization based Functional Link Artificial Neural Network (CSO-FLANN). The comparison of performance is investigated by the Structural Similarity Index (SSIM), Peak Signal to Noise Ratio (PSNR) and Noise Reduction in Decibels (NRDB) values. The simulation results and non-parametric Friedman’s test reveal the superiority of the Jaya-FLMLP filter over others.