2015/09/08 by Andrew Knyazev · 8 citations
Computer Science · Mathematics · #Advanced Graph Neural Networks #Algorithm #Artificial intelligence #Computer science #Discrete mathematics #Graph #Image Enhancement Techniques #Image and Signal Denoising Methods #Laplacian matrix #Mathematics #Noise reduction #Pattern recognition (psychology) #Pixel #Subspace topology #acm:05C85 #acm:68U10 #cs.CV #cs.IT #math.CO #math.IT #msc:05C85 #msc:68U10
paper · pdf · doi:10.1109/globalsip.2015.7418197
published as 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP), Orlando, FL, 14-16 Dec.2015, pp. 260 - 264 · 5 pages; 6 figures. Accepted to IEEE GlobalSIP 2015 conference
arxiv created 2015/09/08 · openalex publication_date 2015/12/01 · arxiv updated 2016/06/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
In [D01:10.1109/ICMEW.2014.6890711], a graph-based denoising is performed by projecting the noisy image to a lower dimensional Krylov subspace of the graph Laplacian, constructed using nonnegative weights determined by distances between image data corresponding to image pixels. We extend the construction of the graph Laplacian to the case, where some graph weights can be negative. Removing the positivity constraint provides a more accurate inference of a graph model behind the data, and thus can improve quality of filters for graph-based signal processing, e.g., denoising, compared to the standard construction, without affecting the costs.