Recursive locally linear mmse motion-compensated image sequence filtering under quantum-limited conditions

Research output: Contribution to journalArticle

3 Scopus citations

Abstract

Quantum noise in image sequences arises in a wide array of applications including medical and astronomical images and remote sensing. It is an undesirable artifact caused by the unavailability or intentional depletion of X-ray and/or light photons necessary for imaging. In this paper, we develop a recursive sliding window, locally linear minimum mean squared error motion-compensated temporal filter for the enhancement of image sequences corrupted by this type of noise. The filter is derived from the formulation of a noise model which describes the underlying physical processes of quantum noise. The recursive implementation of this filter will provide an intuitive manner for addressing subsequent image frames as new observations when acquired under quantum-limited conditions. Experimental results are provided which show the effectiveness of the proposed estimator on both sequences simulated with quantum noise and on real clinical sequences containing natural quantum mottle.

Original languageEnglish (US)
Pages (from-to)349-363
Number of pages15
JournalJournal of Visual Communication and Image Representation
Volume4
Issue number4
DOIs
StatePublished - Dec 1993

ASJC Scopus subject areas

  • Signal Processing
  • Media Technology
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Recursive locally linear mmse motion-compensated image sequence filtering under quantum-limited conditions'. Together they form a unique fingerprint.

  • Cite this