Share Email Print

Proceedings Paper

Training Bayesian networks for image segmentation
Author(s): Xiaojuan Feng; Christopher K. I. Williams
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

We are concerned with the problem of image segmentation in which each pixel is assigned to one of a predefined finite number of classes. In Bayesian image analysis, this requires fusing together local predictions for the class labels with a prior model of segmentations. Markov Random Fields (MRFs) have been used to incorporate some of this prior knowledge, but this not entirely satisfactory as inference in MRFs is NP-hard. The multiscale quadtree model of Bouman and Shapiro (1994) is an attractive alternative, as this is a tree-structured belief network in which inference can be carried out in linear time (Pearl 1988). It is an hierarchical model where the bottom-level nodes are pixels, and higher levels correspond to downsampled versions of the image. The conditional-probability tables (CPTs) in the belief network encode the knowledge of how the levels interact. In this paper we discuss two methods of learning the CPTs given training data, using (a) maximum likelihood and the EM algorithm and (b) conditional maximum likelihood (CML). Segmentation obtained using networks trained by CML show a statistically-significant improvement in performance on synthetic images. We also demonstrate the methods on a real-world outdoor- scene segmentation task.

Paper Details

Date Published: 24 September 1998
PDF: 11 pages
Proc. SPIE 3457, Mathematical Modeling and Estimation Techniques in Computer Vision, (24 September 1998); doi: 10.1117/12.323459
Show Author Affiliations
Xiaojuan Feng, Aston Univ. (United Kingdom)
Christopher K. I. Williams, Aston Univ. (United Kingdom)

Published in SPIE Proceedings Vol. 3457:
Mathematical Modeling and Estimation Techniques in Computer Vision
Francoise J. Preteux; Jennifer L. Davidson; Edward R. Dougherty, Editor(s)

© SPIE. Terms of Use
Back to Top