Plug-and-Play Priors for Bright Field Electron Tomography and Sparse Interpolation

By Suhas Sreehari, S. Venkatakrishnan, Brendt E. Wohlberg, G. Buzzard, Lawrence F. Drummy, Jeff P. Simmons, Charles A. Bouman. IEEE Transactions on Computational Imaging, 2015

373 citations recorded by the source corpus.

Many material and biological samples in scientific imaging are characterized by nonlocal repeating structures. These are studied using scanning electron microscopy and electron tomography. Sparse sampling of individual pixels in a two-dimensional image acquisition geometry, or sparse sampling of projection images with large tilt increments in a tomography experiment, can enable high speed data acquisition and minimize sample damage caused by the electron beam. In this paper, we present an algorithm for electron tomographic reconstruction and sparse image interpolation that exploits the nonlocal redundancy in images. We adapt a framework, termed plug-and-play priors, to solve these imaging problems in a regularized inversion setting. The power of the plug-and-play approach is that it allows a wide array of modern denoising algorithms to be used as a “prior model” for tomography and image interpolation. We also present sufficient mathematical conditions that ensure convergence of the plug-and-play approach, and we use these insights to design a new nonlocal means denoising algorithm. Finally, we demonstrate that the algorithm produces higher quality reconstructions on both simulated and real electron microscope data, along with improved convergence properties compared to other methods.

Subfields: Statistical and Nonlinear Physics, Structural Biology, Surfaces, Coatings and Films

View Paper: https://www.semanticscholar.org/paper/acd22257603757ccb773374774c7a780b85e3a99

How researchers see this paper

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

In Statistical and Nonlinear Physics

In Structural Biology

Lei Zhang

Citations in the corpus

Learning Proximal Operators: Using Denoising Networks for Regularizing Inverse Imaging Problems (2017, background): “Poisson denoising [31], bright field electron tomography [36], super-resolution [2], or hyperspectral image sharpening [38].”
Learning Deep CNN Denoiser Prior for Image Restoration (2017, background): “Some of the other related work can be found in [6, 12, 48, 49, 54, 58].”
Learning to Optimize: A Primer and A Benchmark (2021, background): “The theoretical guarantees of convergence of PnP-ADMM were first provided in [191] under the assumption that the derivative of the denoiser was doubly stochastic.”
Plug-and-Play Methods Provably Converge with Properly Trained Denoisers (2019, methodology): “, 2016), non-local means (Venkatakrishnan et al., 2013; Heide et al., 2014; Sreehari et al., 2016; 2017; Chan, 2019), Gaussian mixture model (Teodoro et al.”
The Little Engine That Could: Regularization by Denoising (RED) (2016, neutral): “which varies through the iterations as well. In terms of convergence, the P3 scheme has been shown to be well-behaved under some conditions on the denoising algorithm used. While the work reported in [31] requires the denoiser to be a symmetric smoothing and non-expansive lter, the later work in [33”
Plug-and-Play ADMM for Image Restoration: Fixed-Point Convergence and Applications (2016, methodology): “This is significan tly weaker than the original method presented in [15], which requires the denoisers to be non-expansive and have symmetr ic gradients [13].”