Filtering of X-Ray Images using Nonlinear Isotropic Diffusion

In this paper an optimization procedure for the number of iterations and two scaling parameters of the Perona-Malik filter is proposed. It is of general type having as aim variables the Peak Signal to Noise Ratio (PSNR) and the Structural Similarity Index Measure (SSIM) of the filtered images. The method is applicable for various types of image content but is extensively tested on X-Ray chest images. The highest PSNR achieved is 35.77 dB and SSIM − 0.9622 when processing images with Additive White Gaussian Noise (AWGN) with a variance of 0.001. One of the aspects for further development of the algorithm is optimizing the time for filtering which turns out to be two orders of magnitude less than that of the Gaussian and the Average filters.