Evaluating diffusion-based image de-noising techniques
Ehsan Nadernejad, Hamid Hassanpour · 2007
This paper evaluates three recently developed diffusion-based image de-noising techniques. Diffusion-based methods have advantage of preserving image edges. In addition, it was shown that the blocking effects and iterations algorithms in these methods are considerably decreased compared to other non-diffusion based approaches. How ever, the threshold value of the diffusion coefficient is a crucial parameter for de-noising technique. Hence this paper suggests the threshold value to be selected adaptively.