Wavelet tree structure based speckle noise removal for optical coherence tomography

Xin Yuan, Xuan Liu, Yang Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We report a new speckle noise removal algorithm in optical coherence tomography (OCT). Though wavelet domain thresholding algorithms have demonstrated superior advantages in suppressing noise magnitude and preserving image sharpness in OCT, the wavelet tree structure has not been investigated in previous applications. In this work, we propose an adaptive wavelet thresholding algorithm via exploiting the tree structure in wavelet coefficients to remove the speckle noise in OCT images. The threshold for each wavelet band is adaptively selected following a special rule to retain the structure of the image across different wavelet layers. Our results demonstrate that the proposed algorithm outperforms conventional wavelet thresholding, with significant advantages in preserving image features.

Original languageEnglish (US)
Title of host publicationOptical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXII
EditorsValery V. Tuchin, Valery V. Tuchin, James G. Fujimoto, Joseph A. Izatt
PublisherSPIE
ISBN (Electronic)9781510614512
DOIs
StatePublished - 2018
EventOptical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXII 2018 - San Francisco, United States
Duration: Jan 29 2018Jan 31 2018

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10483
ISSN (Print)1605-7422

Other

OtherOptical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXII 2018
Country/TerritoryUnited States
CitySan Francisco
Period1/29/181/31/18

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging
  • Biomaterials

Keywords

  • OCT
  • image reconstruction-restoration
  • noise in imaging systems

Fingerprint

Dive into the research topics of 'Wavelet tree structure based speckle noise removal for optical coherence tomography'. Together they form a unique fingerprint.

Cite this