Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
MultiSpectral Probabilistic Diffusion Using Bayesian Classification
 

Summary: Multi­Spectral Probabilistic Diffusion Using
Bayesian Classification
Simon R. Arridge 1 and Andrew Simmons 2
1 Dept. of Computer Science, University College London, Gower Street, London,
WC1E 6BT
2 Dept. Neurology, Institute of Psychiatry, De Crespigny Park, Denmark Hill,
London SE5 8AZ
Abstract. This paper proposes a diffusion scheme for multi­spectral im­
ages which incorporates both spatial derivatives and feature­space clas­
sification. A variety of conductance terms are suggested that use the
posterior probability maps and their spatial derivatives to create resis­
tive boundaries that reflect objectness rather than intensity differences
alone. A theoretical test case is discussed as well as simulated and real
magnetic resonance dual echo images. We compare the method for both
supervised and unsupervised classification.
Keywords: Scale Space, Anisotropic Diffusion, Feature­Space classification,
Magnetic Resonance Imaging.
1 Introduction
Multi­spectral images arise in a number of contexts, either directly, such as RGB
components of a colour image, indirectly, by registration of multiple modalities

  

Source: Arridge, Simon - Department of Computer Science, University College London

 

Collections: Computer Technologies and Information Sciences