MAP Interpolation of an Ising Image Block

This paper considers the problem of finding the set of MAP reconstructions of an \(N\times N\) block conditioned on a boundary configuration consisting of 1 or 2 alternating runs of black and white in a uniform Ising model with no external field. It shows that when the boundary contains a single run, the set of minimum odd bond reconstructions are described by simple paths connecting the endpoints of either the black or white run. When the boundary consists of 2 runs, the set of minimum odd bond reconstructions are formed in one or more of the following ways: by simple paths connecting the endpoints of the two black runs; by simple paths connecting the two white runs; or by three simple paths connecting one of the boundary odd bonds to each of the other three. The paper provides a closed form solution for determining all minimum odd bond reconstructions for a 2-run boundary.

[1]  R. Krishnamoorthi,et al.  Image compression based on a family of stochastic models , 2007, Signal Process..

[2]  Thrasyvoulos N. Pappas.  An adaptive clustering algorithm for image segmentation , 1992, IEEE Trans. Signal Process..

[3]  David L. Neuhoff,et al.  Image interpolation from Manhattan cutset samples via orthogonal gradient method , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[4]  Maurice Queyranne,et al.  On the structure of all minimum cuts in a network and applications , 1982, Math. Program..

[5]  Søren Forchhammer,et al.  Adaptive partially hidden Markov models with application to bilevel image coding , 1999, IEEE Trans. Image Process..

[6]  D. Greig,et al.  Exact Maximum A Posteriori Estimation for Binary Images , 1989 .

[7]  Andrew McCallum,et al.  Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[8]  Matthew G. Reyes,et al.  Row-centric lossless compression of Markov images , 2017, 2017 IEEE International Symposium on Information Theory (ISIT).

[9]  Dongming Zhao,et al.  Multiresolution image compression using Gaussian Markov random fields , 1993, Proceedings of 36th Midwest Symposium on Circuits and Systems.

[10]  Aleksandra Pizurica,et al.  Context-Aware Patch-Based Image Inpainting Using Markov Random Field Modeling , 2015, IEEE Transactions on Image Processing.

[11]  Matthew G. Reyes,et al.  Cutset width and spacing for Reduced Cutset Coding of Markov random fields , 2016, 2016 IEEE International Symposium on Information Theory (ISIT).

[12]  Mihalis Yannakakis,et al.  Suboptimal Cuts: Their Enumeration, Weight and Number (Extended Abstract) , 1992, ICALP.

[13]  R. Baxter.  Exactly solved models in statistical mechanics , 1982 .

[14]  Matthew G. Reyes,et al.  Lossy Cutset Coding of Bilevel Images Based on Markov Random Fields , 2014, IEEE Transactions on Image Processing.