Texture synthesis and unsupervised recognition with a nonparametric multiscale Markov random field model

We present noncausal, nonparametric, multiscale, Markov random field model for synthesising and recognising texture. The model has the ability to capture the characteristics of a wide variety of textures. For texture synthesis, we use our own novel multiscale approach, incorporating local annealing,...

Full description

Saved in:
Bibliographic Details
Main Authors: Paget, R., Longstaff, I.D.
Format: Conference Proceeding
Language:English
Subjects:
Image segmentation
Image texture analysis
Information processing
Libraries
Markov random fields
Reactive power
Read only memory
Signal processing
Signal synthesis
Testing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present noncausal, nonparametric, multiscale, Markov random field model for synthesising and recognising texture. The model has the ability to capture the characteristics of a wide variety of textures. For texture synthesis, we use our own novel multiscale approach, incorporating local annealing, allowing one to use large neighbourhood systems to model some complex textures. We show how one is able to manipulate the statistical order of our high dimensional model without over compromising the integrity of the representation. Also, by varying the statistical order of our model we are able to optimise it for the unsupervised recognition of textures with respect to textures that have not been modelled.
ISSN:1051-4651
2831-7475
DOI:10.1109/ICPR.1998.711876