Paper Info
Title
Bayesian Non-Parametric Image Segmentation With Markov Random Field Prior
Abstract
In this paper, a Bayesian framework for non-parametric density estimation with spatial smoothness constraints is presented for image segmentation. Unlike common parametric methods such as mixtures of Gaussians, the proposed method does not make strict assumptions about the shape of the density functions and thus, can handle complex structures. The multiclass kernel density estimation is considered as an unsupervised learning problem. A Dirichlet compound multinomial (DCM) prior is used to model the class label prior probabilities and a Markov random field (MRF) is exploited to impose the spatial smoothness and control the confidence on the Dirichlet hyper-parameters, as well. The proposed model results in a closed form solution using an expectation-maximization (EM) algorithm for maximum a posteriori (MAP) estimation. This provides a huge advantage over other models which utilize more complex and time consuming methods such as Markov chain Monte Carlo (MCMC) or variational approximation methods. Several experiments on natural images are performed to present the performance of the proposed model compared to other parametric approaches.
Year
DOI
Venue
2013
10.1007/978-3-642-38886-6_8
IMAGE ANALYSIS, SCIA 2013: 18TH SCANDINAVIAN CONFERENCE
Keywords
Field
DocType
Multiclass kernel density estimation, Dirichlet compound multinomial distribution, Markov random field prior, Image segmentation
Density estimation,Pattern recognition,Markov chain Monte Carlo,Markov random field,Computer science,Image segmentation,Parametric statistics,Artificial intelligence,Dirichlet distribution,Maximum a posteriori estimation,Kernel density estimation
Conference
Volume
ISSN
Citations 
7944
0302-9743
0
PageRank 
References 
Authors
0.34
7
1
Name
Order
Citations
PageRank
Ehsan Amid1216.83