We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.NA

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Numerical Analysis

Title: Plug-and-Play gradient-based denoisers applied to CT image enhancement

Abstract: Blur and noise corrupting Computed Tomography (CT) images can hide or distort small but important details, negatively affecting the diagnosis. In this paper, we present a novel gradient-based Plug-and-Play algorithm, constructed on the Half-Quadratic Splitting scheme, and we apply it to restore CT images. In particular, we consider different schemes encompassing external and internal denoisers as priors, defined on the image gradient domain. The internal prior is based on the Total Variation functional. The external denoiser is implemented by a deep Convolutional Neural Network (CNN) trained on the gradient domain (and not on the image one, as in state-of-the-art works). We also prove a general fixed-point convergence theorem under weak assumptions on both internal and external denoisers. The experiments confirm the effectiveness of the proposed framework in restoring blurred noisy CT images, both in simulated and real medical settings. The achieved enhancements in the restored images are really remarkable, if compared to the results of many state-of-the-art methods.
Comments: Submitted to journal
Subjects: Numerical Analysis (math.NA); Computer Vision and Pattern Recognition (cs.CV)
MSC classes: 65K10, 65R30, 68U10, 68T07, 65F22
ACM classes: G.1.6; G.1.10; I.4.3; I.4.4; I.4.5; I.4.6; I.2.6; I.2.0; I.4.3
Cite as: arXiv:2102.07510 [math.NA]
  (or arXiv:2102.07510v2 [math.NA] for this version)

Submission history

From: Andrea Sebastiani [view email]
[v1] Mon, 15 Feb 2021 12:19:28 GMT (5705kb,D)
[v2] Fri, 19 Mar 2021 10:10:08 GMT (4364kb,D)

Link back to: arXiv, form interface, contact.