2011 | OriginalPaper | Buchkapitel
Kaiser Window Function Non-Local Means for Image Denoising
verfasst von : Hongwei Li, Pu Zhang, Yuan Huang
Erschienen in: Informatics in Control, Automation and Robotics
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Kernel function has a great influence on the results denoised by Non-Local Means (NLM), the original NLM method uses the gaussian kernel function to product the patch pixels, and the central pixel’s 4-neighborhood and its 8-neighborhood are given the same kernel value. While in the fact that different neighborhood should be given different kernel values, and in this case, we proposed an improved NLM algorithm based on Kaiser Window function, in which the central pixel’s different scale neighborhood has been assigned different values from the Kaiser Window function. Through a myriad of experiments, both visual quality and PSNR proved that our method is better than the original algorithm.