Segmenting Images Corrupted by Correlated Noise

Authors:
Thomas C. M. Lee
Affiliations:
Univ. of Chicago, Chicago, IL
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
1998

Citing 8
Cited 2

Segmentation through Variable-Order Surface Fitting

IEEE Transactions on Pattern Analysis and Machine Intelligence
Hierarchy in Picture Segmentation: A Stepwise Optimization Approach

IEEE Transactions on Pattern Analysis and Machine Intelligence
Estimation of Planar Curves, Surfaces, and Nonplanar Space Curves Defined by Implicit Equations with Applications to Edge and Range Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Bayesian Segmentation Methodology for Parametric Image Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Image analysis for the biological sciences

Image analysis for the biological sciences
Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Model-Fitting Approach to Cluster Validation with Application to Stochastic Model-Based Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Seeded Region Growing

IEEE Transactions on Pattern Analysis and Machine Intelligence

Asymptotic bootstrap corrections of AIC for linear regression models

Signal Processing
Multi-component image segmentation in homogeneous regions based on description length minimization: Application to speckle, Poisson and Bernoulli noise

Pattern Recognition

Quantified Score

Hi-index	0.15

Visualization

Abstract

Image segmentation is fundamental to many image analysis problems. It aims to partition a digital image into a set of nonoverlapping homogeneous regions. The main contribution of this paper is the development of a new segmentation procedure which is designed to segment images corrupted by correlated noise. This new segmentation procedure is based on Rissanen's minimum description length (MDL) principle and consists of two components: i) an MDL-based criterion in which the "best" segmentation is defined as its minimizer and ii) a merging algorithm which attempts to locate this minimizer. The performance of this procedure is illustrated via a simulation study, with promising results.