Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Poisson Image Deconvolution by a Plug-and-Play Quantum Denoising Scheme

Sayantan Dutta 1, 2 Adrian Basarab 2 Bertrand Georgeot 1, 3 Denis Kouamé 2
2 IRIT-MINDS - CoMputational imagINg anD viSion
IRIT - Institut de recherche en informatique de Toulouse
3 Information et Chaos Quantiques (LPT)
LPT - Laboratoire de Physique Théorique
Abstract : This paper introduces a new Plug-and-Play (PnP) alternating direction of multipliers (ADMM) scheme based on a recently proposed denoiser using the Schroedinger equation solutions of quantum physics. The proposed algorithm referred to as QAB-PnP is well-adapted to Poisson noise, which is very common for imaging applications, such as, limited photon acquisition. In contrast to existing PnP approaches using a variance stabilizing transformation that is not invariant to deconvolution operation, the proposed method does not suffer from this theoretical problem. Moreover, numerical results show the superiority of the proposed scheme compared to recent state-of-the-art techniques, for both low and high signal-to-noise-ratio scenarios.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02972150
Contributor : Bertrand Georgeot <>
Submitted on : Tuesday, October 20, 2020 - 10:44:34 AM
Last modification on : Monday, November 16, 2020 - 11:16:43 AM

Links full text

Identifiers

  • HAL Id : hal-02972150, version 1
  • ARXIV : 2010.09321

Citation

Sayantan Dutta, Adrian Basarab, Bertrand Georgeot, Denis Kouamé. Poisson Image Deconvolution by a Plug-and-Play Quantum Denoising Scheme. 2020. ⟨hal-02972150⟩

Share

Metrics

Record views

21