From Local Kernel to Nonlocal Multiple-Model Image Denoising

By Vladimir Katkovnik, Alessandro Foi, Karen O. Egiazarian, Jaakko T. Astola. International Journal of Computer Vision, 2009

408 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Geophysics

View Paper: https://www.semanticscholar.org/paper/6f5008eb780fcf4afd74504c026a8cc0733342e6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Geophysics

Citations in the corpus

Sparse Modeling for Image and Vision Processing (2014, methodology): “Another link between neural networks and dictionary learning has been established by Kavukcuoglu et al. [2010b] and Gregor and LeCun [2010], who have trained neural networks to approximate sparse codes.”
From Heuristic Optimization to Dictionary Learning: A Review and Comprehensive Comparison of Image Denoising Algorithms (2014, background): “[45] classified denoising filters according to local/nonlocal and pointwise/multipoint.”
A supervised patch-based approach for human brain labeling (2011, positive): “mentioned in [17] (“in multipoint image estimation a weighted”
BM3D Frames and Variational Image Deblurring (2011, background): “Its detailed discus sion can be found in [6].”
Video Denoising, Deblocking, and Enhancement Through Separable 4-D Nonlocal Spatiotemporal Transforms (2012, methodology): “Until now, the most effective image-denoising algorithm, termed BM3D [6], [10], relies on the so-called grouping and collaborative filtering paradigm: at first mutually similar 2-D image blocks are stacked into a 3-D group (grouping), then the group is filtered through a transform-domain shrinkage (”
Guided Image Filtering (2010, background): “denoising (see [53]) and is a building block of the very”
Statistics of Patch Offsets for Image Completion (2012, positive): “Gradient-domain statistics have been applied in denoising [29], deconvolution [30], and inpainting [3], whereas recent patch-domain statistics have been shown very successful in denoising [29] and super-resolution [31].”
Denoising of 3D magnetic resonance images by using higher-order singular value decomposition (2015, methodology): “This method combines nonlocal and transform-domain approaches (Katkovnik et al., 2010) and presents an effective denoising framework by grouping similar patches into a 3D array, then filtering the 3D array by using sparse representation in the transform domain, and finally aggregating multiple estim”
Patch-Based Nonlocal Functional for Denoising Fluorescence Microscopy Image Sequences (2010, methodology): “Many methods have been described in the literature, and a recent comprehensive review can be found in [1], [2].”
Nonlocally Centralized Sparse Representation for Image Restoration (2013, methodology): “From left to right and top to bottom: original image, noisy image (σ = 100), and denoised images by SAPCA-BM3D [39] (PSNR = 25.20 dB; FSIM = 0.8065), LSSC [18] (PSNR = 25.63 dB; FSIM = 0.8017), EPLL [33] (PSNR = 25.44 dB; FSIM = 0.8100), and NCSR (PSNR = 25.65 dB; FSIM = 0.8068). achieves highly…”
Complex diffusion-weighted image estimation via matrix recovery under general noise models (2018, neutral): “However, the estimates could benefit from more refined locally adaptive criteria to build the spatial patches that also take into account the signal similarities (Katkovnik et al., 2010) as well as from patches being employed both in the spatial and the diffusion coordinates.”
NL-SAR: A Unified Nonlocal Framework for Resolution-Preserving (Pol)(In)SAR Denoising (2015, background): “age processing now derive from this idea [28], including BM3D considered as one of the most powerful approaches [29].”
A Nonlocal SAR Image Denoising Algorithm Based on LLMMSE Wavelet Shrinkage (2012, background): “At present, BM3-D can be arguably [27] considered the state of the art for AWGN denoising.”
Image Completion Approaches Using the Statistics of Similar Patches (2014, methodology): “Patch-domain statistics have been shown very successful in denoising [44] and super-resolution [46].”
Denoising of 3 D Magnetic Resonance Images by Using 1 Higher-Order Singular Value Decomposition 2 (2014, methodology): “This method combines nonlocal and 29 transform-domain approaches (Katkovnik et al., 2010) and presents an effective 30 5 denoising framework by grouping similar patches into a 3D array, then filtering the 1 3D array by using sparse representation in the transform domain, and finally 2 aggregating…”
Iterative Weighted Maximum Likelihood Denoising With Probabilistic Patch-Based Weights (2009, background): “[29] for a deeper analysis of the connections and evolutions of”
Diffusion Weighted Image Denoising Using Overcomplete Local PCA (2013, background): “However, it can be estimated theoretically and inverted in the original domain using the properties of the first moment of a Rician distribution [38].”
Poisson Noise Reduction with Non-local PCA (2012, methodology): “Since the introduction of non-local methods for image denoising [8], these methods have proved to outperform previously considered approaches [1, 11, 12, 30] (extensive comparisons of recent denoising method can be found for Gaussian noise in [21, 26]).”