An analytical method for the adaptive computation of threshold of gradient modulus in 2D anisotropic diffusion filter

摘要

In spite of the extensive application of Anisotropic Diffusion (AD) filter in software packages for medical image analysis, denoising and edge preservation offered by it depends exclusively on the selection of the value of Threshold of Gradient Modulus (TGM). Tuning the TGM to its optimum value through trial and error is subjective and tiring. An analytical model to compute the optimum value of TGM adaptively from the mean gradient of the image itself is proposed in this article. The qualitative examination of the gradient and true edge maps of the original and restored Magnetic Resonance images revealed that analytically computed TGM ensures best trade-off between noise suppression and edge preservation. (C) 2017 Nalecz Institute of Biocybemetics and Biomedical Engineering of the Polish Academy of Sciences. Published by Elsevier B.V. All rights reserved.