Region-adaptive transform based on a stochastic model

Author

Stiller, Christoph ; Konrad, Janusz é

Author_Institution

INRS Telecommun., Ile des Soeurs, Que., Canada

Volume

2

fYear

1995

fDate

23-26 Oct 1995

Firstpage

264

Abstract

This paper is concerned with linear transforms for arbitrarily-shaped image segments. In contrast to other techniques described in the literature, the proposed transform is based upon a stochastic model of image covariance within the considered region. Emerging from a separable stationary Markov model proposed for rectangular regions, we derive a non-stationary Markov model with natural boundary conditions. We compute its eigentransform, which is the optimum linear transform under a broad variety of performance measures. For the special case of a rectangular region, the method yields the DCT basis functions. Simulation results for natural imagery are provided

Keywords

Markov processes; adaptive signal processing; covariance analysis; discrete cosine transforms; eigenvalues and eigenfunctions; image coding; image segmentation; stochastic processes; transform coding; DCT basis functions; arbitrarily shaped image segments; eigentransform; image coding; image covariance; linear transforms; natural boundary conditions; natural imagery; nonstationary Markov model; optimum linear transform; performance measures; rectangular regions; region adaptive transform; separable stationary Markov model; simulation results; stochastic model; Boundary conditions; Business; Discrete cosine transforms; Discrete transforms; Image coding; Image segmentation; Region 7; Shape; Stochastic processes; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 1995. Proceedings., International Conference on

Conference_Location

Washington, DC

Print_ISBN

0-8186-7310-9

Type

conf

DOI

10.1109/ICIP.1995.537465

Filename

537465