TY - JOUR
T1 - The steerable graph Laplacian and its application to filtering image datasets
AU - Landa, Boris
AU - Shkolnisky, Yoel
N1 - Publisher Copyright:
© 2018 Society for Industrial and Applied Mathematics.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2018
Y1 - 2018
N2 - In recent years, improvements in various image acquisition techniques gave rise to the need for adaptive processing methods, aimed particularly for large datasets corrupted by noise and deformations. In this work, we consider datasets of images sampled from a low-dimensional manifold (i.e., an image-valued manifold), where the images can assume arbitrary planar rotations. To derive an adaptive and rotation-invariant framework for processing such datasets, we introduce a graph Laplacian (GL)-like operator over the dataset, termed a steerable graph Laplacian. Essentially, the steerable GL extends the standard GL by accounting for all (infinitely many) planar rotations of all images. As it turns out, similarly to the standard GL, a properly normalized steerable GL converges to the Laplace--Beltrami operator on the low-dimensional manifold. However, the steerable GL admits an improved convergence rate compared to the GL, where the improved convergence behaves as if the intrinsic dimension of the underlying manifold is lower by one. Moreover, it is shown that the steerable GL admits eigenfunctions of the form of Fourier modes (along the orbits of the images' rotations) multiplied by eigenvectors of certain matrices, which can be computed efficiently by the FFT. For image datasets corrupted by noise, we employ a subset of these eigenfunctions to ``filter"" the dataset via a Fourier-like filtering scheme, essentially using all images and their rotations simultaneously. We demonstrate our filtering framework by denoising simulated single-particle cryo-electron-microscopy image datasets.
AB - In recent years, improvements in various image acquisition techniques gave rise to the need for adaptive processing methods, aimed particularly for large datasets corrupted by noise and deformations. In this work, we consider datasets of images sampled from a low-dimensional manifold (i.e., an image-valued manifold), where the images can assume arbitrary planar rotations. To derive an adaptive and rotation-invariant framework for processing such datasets, we introduce a graph Laplacian (GL)-like operator over the dataset, termed a steerable graph Laplacian. Essentially, the steerable GL extends the standard GL by accounting for all (infinitely many) planar rotations of all images. As it turns out, similarly to the standard GL, a properly normalized steerable GL converges to the Laplace--Beltrami operator on the low-dimensional manifold. However, the steerable GL admits an improved convergence rate compared to the GL, where the improved convergence behaves as if the intrinsic dimension of the underlying manifold is lower by one. Moreover, it is shown that the steerable GL admits eigenfunctions of the form of Fourier modes (along the orbits of the images' rotations) multiplied by eigenvectors of certain matrices, which can be computed efficiently by the FFT. For image datasets corrupted by noise, we employ a subset of these eigenfunctions to ``filter"" the dataset via a Fourier-like filtering scheme, essentially using all images and their rotations simultaneously. We demonstrate our filtering framework by denoising simulated single-particle cryo-electron-microscopy image datasets.
KW - Graph Laplacian
KW - Manifold denoising
KW - Manifold learning
KW - Rotation invariance
KW - Steerable filters
UR - http://www.scopus.com/inward/record.url?scp=85064230143&partnerID=8YFLogxK
U2 - 10.1137/18M1169394
DO - 10.1137/18M1169394
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:85064230143
SN - 1936-4954
VL - 11
SP - 2254
EP - 2304
JO - SIAM Journal on Imaging Sciences
JF - SIAM Journal on Imaging Sciences
IS - 4
ER -