Arak-Clifford-Surgailis Tessellations. Basic Properties and Variance of the Total Edge Length
Non-homogeneous random tessellations in the plane with T-shaped vertices are considered, which are defined as Gibbsian modifications of a Poisson line process with arbitrary locally finite intensity measure. In the homogeneous set-up they can be regarded as a specific case of the general Arak-Surgailis polygonal fields in the plane and share some properties with the iteration stable (STIT) tessellations. In the non-homogeneous and anisotropic environment an explicit expression for the partition function is provided and first- and second-order properties of the random length measure of the tessellations restricted to a convex sampling window are derived. In the more special isotropic regime a closed formula for the pair-correlation function and the variance of the total edge length is obtained.
KeywordsGibbs modification Pair-correlation function Polygonal Markov field Random tessellation Stochastic geometry
- 1.Arak, T.: On Markovian random fields with finite number of values. In: 4th USSR-Japan Symposium on Probability Theory and Mathematical Statistics. Abstracts of Communications, Tbilisi (1982) Google Scholar
- 5.Clifford, P., Nicholls, G.: A Metropolis sampler for polygonal image reconstruction. Available online at http://www.stats.ox.ac.uk/clifford/papers/met_poly.html (1994)
- 7.Kluszczyński, R., van Lieshout, M.N.M., Schreiber, T.: Image segmentation by polygonal Markov fields. Ann. Inst. Stat. Math. 59, 615–625 (2007) Google Scholar
- 12.Schreiber, T., Thäle, C.: Second-order theory for iteration stable tessellations. arXiv:1103.3959 [math.PR] (2011)