Discrete Tomography pp 297-316 | Cite as

# Reconstruction of Binary Images via the EM Algorithm

## Abstract

The problem of reconstructing a binary function x defined on a finite subset of a lattice ℤ, from an arbitrary collection of its partial sums is considered. The approach is based on (a) relaxing the binary constraints
\(x\left( i \right) = 0\)
or 1 to interval constraints \(0x\left( i \right)1,i \in {\Bbb Z}\),and (b) applying a minimum distance method (using Kullback-Leibler’s information divergence index as our distance function) to find such an *x* — say, \(\hat x\)— for which the distance between the observed and the theoretical partial sums is as small as possible (Turning this \(\hat x\) into a binary function can be done as a separate postprocessing step: for instance,through thresholding, or through some additional Bayes modeling.) This minimum-distance solution is derived via a new EM algorithm that extends the often-studied EM/maximum likelihood (EM/ML) algorithm in emission tomography and certain linear-inverse problems to include lower-and upper-bound constraints on the function x. Properties of the algorithm including convergence and uniqueness conditions on the solution (or parts of it) are described.

## Keywords

Binary Image Maximum Likelihood Estimator Probability Vector Binary Function Binary Constraint## Preview

Unable to display preview. Download preview PDF.

## References

- [1]Y. Vardi and D. Lee, “The discrete Radon transform and its approximate inversion via the EM algorithm,” International J. Imaging Systems and Techn. 9, 155–173 (1998).CrossRefGoogle Scholar
- [2]Y. Vardi and C.-H. Zhang, “Estimating mixing probabilities subject to lower and upper bound constraints with applications,” Technical Report #97–006, Department of Statistics, Rutgers University, Piscataway, NJ (1997).Google Scholar
- [3]
*U. Rothblum and Y. Vardi*“Maximum likelihood estimation of cell probabilities in constrained multinomial models,” J. Statist. Computation and Simulation, 61, 141–161 (1998).CrossRefGoogle Scholar - [4]
*A. P. Dempster, N. M. Laird, and R. B. Rubin*“Maximum likelihood from incomplete data via the EM algorithm (with Discussion),” J. R. Statist. Soc. B 39, 1–38 (1977).Google Scholar - [5]
*G. T. Herman*“Reconstruction of binary patterns from a few projections,” In A. Günther, B. Levrat, and H. Lipps, International Computing Symposium 1973,*(North-Holland Publ. Co., Amsterdam)*pp. 371–378,1974.Google Scholar - [6]L. A. Shepp, and Y. Vardi, “Maximum likelihood reconstruction for emission tomography,” IEEE Trans. Med. Imaging 1, 113–122 (1982).CrossRefPubMedGoogle Scholar
- [7]
*Y. Vardi, L. A. Shepp and L. Kaufman*“A statistical model for positron emission tomography (with Discussion),” J. Amer. Statist. Assoc. 80, 8–37 (1985).CrossRefGoogle Scholar - [8]Y. Vardi and D. Lee, “From image deblurring to optimal investments: Maximum likelihood solutions for positive linear inverse problems (with Discussion),” J. R. Statist. Soc. B 55, 569–612 (1993).Google Scholar
- [9]
*T. M. Cover*“An algorithm for maximizing expected log investment return,” IEEE Trans. Inform. Theory 30, 369–373 (1984).CrossRefGoogle Scholar - [10]I. Csiszár and G. Tusnády, “Information geometry and alternating minimization procedures,” Statist. Decis., Suppl. 1, 205–237 (1984).Google Scholar
- [11]P. Fishburn, P. Schwander, L. Shepp, and R. Vanderbei, “The discrete Radon transform and its approximate inversion via linear programming,” Discrete Applied Mathematics 75, 39–61 (1997).CrossRefGoogle Scholar
- [12]R. Gordon, R. Bender, and G. T. Herman, “Algebraic reconstruction techniques (ART) for three-dimensional electron microscopy and x-ray photography,” J. Theoret. Biol. 29, 471–481 (1970).CrossRefGoogle Scholar
- [13]
*L. M. Bregman*“The relaxation method of finding the common point of convex sets and its application to the solution of problems in convex programming,” U.S.S.R. Comput. Math. and Math. Phys. 7, 200–217 (1967).CrossRefGoogle Scholar - [14]Y. Censor and S. A. Zenios, Parallel Optimization: Theory, Algorithms, and Applications (Oxford University Press, New York), 1997.Google Scholar
- [15]Y. Censor, A. R. De Pierro, T. Elfving, G. T. Herman, and A. N. Iusem, “On iterative methods for linearly constrained entropy maximization,”
*In A. Wakulicz, Numerical Analysis and Mathematical Modeling, (PWN - Polish Scientific Publishers, Warsaw)*pp. 145–163, 1990.Google Scholar - [16]
*Y. Censor*“Finite series-expansion reconstruction methods,” Proc. IEEE 71, 409–419 (1983).CrossRefGoogle Scholar - [17]
*G. T. Herman*“Applications to maximum entropy and Bayesian optimization methods to image reconstruction from projections*” In C. P. Smith and W. T. Grandy, Jr., Maximum-Entropy and Bayesian Methods in Inverse Problems, (D. Reidel Publishing Company, Dordrecht*The Netherlands), pp. 319–338, 1985.CrossRefGoogle Scholar - [18]C. L. Byrne, “Iterative image reconstruction algorithms based on cross-entropy minimization,” IEEE Trans. on Image Processing 2, 96103 (1993).CrossRefGoogle Scholar
- [19]R. Aharoni, G. T. Herman, and A. Kuba, “Binary vectors partially determined by linear equation systems,” Discrete Math. 171, 1–16 (1997).CrossRefGoogle Scholar
- [20]M. T. Chan, G. T. Herman, and E. Levitan, “Bayesian image reconstruction using image-modeling Gibbs priors,” Int. J. Imaging Sys. Tech. 9, 85–98 (1997).CrossRefGoogle Scholar
- [21]S-K. Chang, “The reconstruction of binary patterns from their projections,” Comm. of the ACM 14, 21–25 (1971).CrossRefGoogle Scholar
- [22]
*H. W. Kuhn and A. W. Tucker*“Nonlinear programming.” Proc. Second Berkeley Symp. Math. Statist. Probab. 1, 481–492 (1951).Google Scholar