Neighbour-based MLPs

Author

Dunne, R.A. ; Campbell, N.A.

Author_Institution

Victoria Univ. of Technol., Melbourne, Vic., Australia

Volume

1

fYear

1995

fDate

Nov/Dec 1995

Firstpage

270

Abstract

This paper reviews a Bayesian approach to classifying multi-spectral image data where the pixel labels are assumed to be spatially correlated. A Markov random field (MRF) model is introduced to model localized dependence, so that a label is assumed to be conditional on the labels of the neighbouring pixels only. The multi-layer perceptron model is extended to incorporate the MRF model and it is shown that the posterior distribution of labels given spectral values and neighbouring labels can be maximized by an iterative updating procedure. This updating procedure is in fact an implementation of a (modified) Hopfield network. Finally an example is presented that consists of classifying a remotely sensed image of an agricultural property

Keywords

Bayes methods; Hopfield neural nets; Markov processes; decision theory; image classification; iterative methods; multilayer perceptrons; remote sensing; Bayesian approach; Hopfield network; Markov random field model; agricultural property; iterative updating procedure; localized dependence; multi-spectral image data; neighbour-based multilayer perceptrons; posterior distribution; remotely sensed image; Australia; Bayesian methods; Context modeling; Markov random fields; Mathematics; Multilayer perceptrons; Multispectral imaging; Pixel; Remote sensing; Statistics;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1995. Proceedings., IEEE International Conference on

Conference_Location

Perth, WA

Print_ISBN

0-7803-2768-3

Type

conf

DOI

10.1109/ICNN.1995.488107

Filename

488107