SUSAN—A New Approach to Low Level Image Processing

  • Authors:
  • Stephen M. Smith;J. Michael Brady

  • Affiliations:
  • Department of Clinical Neurology, Oxford University, Oxford, UK. E-mail: steve@robots.ox.ac.uk;Department of Engineering Science, Oxford University, Oxford, UK. E-mail: jmb@robots.ox.ac.uk

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 1997

Quantified Score

Hi-index 0.02

Visualization

Abstract

This paper describes a new approach to low level image processing; inparticular, edge and corner detection and structure preserving noisereduction.Non-linear filtering is used to define which parts of the image areclosely related to each individual pixel; each pixel has associated with it alocal image region which is of similar brightness to that pixel. The newfeature detectors are based on the minimization of this local image region,and the noise reduction method uses this region as the smoothingneighbourhood. The resulting methods are accurate, noise resistant andfast.Details of the new feature detectors and of the new noise reductionmethod are described, along with test results.