Image segmentation by polygonal Markov Fields

This paper advocates the use of multi-coloured polygonal Markov fields for model-based image segmentation. The formal construction of consistent multi-coloured polygonal Markov fields by Arak–Clifford–Surgailis and its dynamic representation are specialised and adapted to our context. We then formulate image segmentation as a statistical estimation problem for a Gibbsian modification of an underlying polygonal Markov field, and discuss the choice of Hamiltonian. Monte Carlo techniques, including novel Gibbs updates for the Arak model, to estimate the model parameters and find an optimal partition of the image are developed. The approach is applied to image data, the first published application of polygonal Markov fields to segmentation problems in the mathematical literature.

[1]  T. Liggett.  Interacting Particle Systems , 1985 .

[2]  Sebastian Thrun,et al.  Robotic Mapping with Polygonal Random Fields , 2005, UAI.

[3]  D. Surgailis,et al.  Markov fields with polygonal realizations , 1989 .

[4]  Donatas Surgailis,et al.  Consistent polygonal fields , 1991 .

[5]  Peter Green,et al.  Highly Structured Stochastic Systems , 2003 .

[6]  Leszek Wojnar,et al.  Image Analysis , 1998 .

[7]  Vivek S. Borkar,et al.  Stochastic processes that generate polygonal and related random fields , 1996, IEEE Trans. Inf. Theory.

[8]  Mathias Ortner,et al.  Processus ponctuels marqués pour l'extraction automatique de caricatures de bâtiments à partir de modèles numériques d'élévation , 2004 .

[9]  Peter Clifford,et al.  Reconstruction of polygonal images , 1989 .

[10]  Gerhard Winkler,et al.  Image Analysis, Random Fields and Markov Chain Monte Carlo Methods: A Mathematical Introduction , 2002 .

[11]  Håkon Tjelmeland,et al.  Markov Random Fields with Higher‐order Interactions , 1998 .

[12]  Georgy L. Gimel'farb,et al.  Image Textures and Gibbs Random Fields , 1999, Computational Imaging and Vision.

[13]  P. Clifford,et al.  Point-based polygonal models for random graphs , 1993, Advances in Applied Probability.

[14]  van Marie-Colette Lieshout.  Interview about work wit R. Klusczynski and T. Schreiber on image segmentation by Polygonal Markov fields , 2005 .

[15]  Random dynamics and thermodynamic limits for polygonal Markov fields in the plane , 2005, Advances in Applied Probability.

[16]  Jesper Møller,et al.  Markov connected component fields , 1996, Advances in Applied Probability.

[17]  H. Haario,et al.  Simulated annealing process in general state space , 1991, Advances in Applied Probability.

[18]  Marie-Colette van Lieshout,et al.  An Algorithm for Binary Image Segmentation Using Polygonal Markov Fields , 2005, ICIAP.

[19]  J. Heikkinen,et al.  Non‐parametric Bayesian Estimation of a Spatial Poisson Intensity , 1998 .

[20]  Håvard Rue,et al.  Advances in Bayesian Image Analysis , 2003 .

[21]  Adrian Baddeley,et al.  ICM for Object Recognition , 1992 .

[22]  Azriel Rosenfeld,et al.  Digital Picture Processing , 1976 .

[23]  S. Osher,et al.  Geometric Level Set Methods in Imaging, Vision, and Graphics , 2011, Springer New York.

[24]  Geoff K. Nicholls,et al.  Bayesian image analysis with Markov chain Monte Carlo and coloured continuum triangulation models , 1998 .

[25]  Øivind Skare,et al.  Coloured Voronoi tessellations for Bayesian image analysis and reservoir modelling , 2001 .

[26]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .