Bayesian approach to inverse scattering with topological priors

Thumbnail Image
Full text at PDC
Publication Date
Advisors (or tutors)
Journal Title
Journal ISSN
Volume Title
IOP Publishing
Google Scholar
Research Projects
Organizational Units
Journal Issue
We propose a Bayesian inference framework to estimate uncertainties in inverse scattering problems. Given the observed data, the forward model and their uncertainties, we find the posterior distribution over a finite parameter iield representing the objects. To construct the prior distribution we use a topological sensitivity analysis. We demonstrate the approach on the Bayesian solution of 2D inverse problems in light and acoustic holography with synthetic data. Statistical information on objects such as their center location, diameter size, orientation, as well as material properties, are extracted by sampling the posterior distribution. Assuming the number of objects known, comparison of the results obtained by Markov Chain Monte Carlo sampling and by sampling a Gaussian distribution found by linearization about the maximum a posteriori estimate show reasonable agreement. The latter procedure has low computational cost, which makes it an interesting tool for uncertainty studies in 3D. However, MCMC sampling provides a more complete picture of the posterior distribution and yields multi-modal posterior distributions for problems with larger measurement noise. When the number of objects is unknown, we devise a stochastic model selection framework.
[1] C.Y. Ahn, K. Jeon, Y.K. Ma, W.K. Park, A study on the topological derivative-based imaging of thin electromagnetic inhomogeneities in limited-aperture problems, Inverse Probl. 30 (2014), 105004. [2] K. Belkebir, M. Saillard, Testing inversion algorithms against experimental data, Inverse Probl. 17 (2001) 1565-1571. [3] C.M. Bishop, Pattern Recognition and Machine learning, Springer 2006. [4] C. Borges, A. Gillman, L. Greengard. High resolution inverse scattering in two dimensions using recursive linearization, SIAM J. Imaging Sciences, 10 (2017), 641-664. [5] G.E.P. Box, G.C. Tiao, Multiparameter problem from a Bayesian point of view, The Annals of Mathematical Statistics, 36 (1965), 1468-1482. [6] T. Bui-Thanh, O. Ghattas, J. Martin, G. Stadler, A computational framework for in�nitedimensional Bayesian inverse problems. Part I: The linearized case, with application to global seismic inversion. SIAM Journal on Scienti�c Computing 35 (2013), A2494-A2523. [7] T. Bui-Thanh, O. Ghattas, An analysis of in�nite dimensional Bayesian inverse shape acoustic scattering and its numerical approximation, SIAM/ASA Journal on Uncertainty Quanti�cation 2 (2014) 203-222. [8] F. Cakoni, D. Colton, P. Monk, The Linear Sampling Method in Inverse Electromagnetic Scattering, SIAM, 2011. [9] B.P. Carlin, T.L. Louis, Bayes and Empirical Bayes Methods for Data Analysis (2nd ed.). Chapman & Hall/CRC, 2000. 10] A. Carpio, M.L. Rap�un, Topological derivatives for shape reconstruction, Lecture Notes in Mathematics 1943 (2008), 85-133. [11] A. Carpio, M.L. Rap�un, Solving inverse inhomogeneous problems by topological derivative methods, Inverse Problems 24 (2008) 045014. [12] A. Carpio, T.G. Dimiduk, V. Selgas, P. Vidal, Optimization methods for in-line holography, SIAM Journal on Imaging Sciences 11 (2018), 923-956. [13] A. Carpio, T.G. Dimiduk, F. Le Lou�er, M.L. Rap�un, When topological derivatives met regularized Gauss-Newton iterations in holographic 3D imaging, J. Comp. Phys. 388 (2019), 224-251. [14] F. Caubet, M. Godoy, C. Conca, On the detection of several obstacles in 2D Stokes ow: topological sensitivity and combination with shape derivatives, Inverse Probl. Imaging 10 (2016) 327-367. [15] H. Chipman, E.I. George, R.E. McCulloch, The practical implementation of Bayesian model selection, in Model Selection IMS Lecture Notes - Monograph Series (2001) Volume 38, 66-134. [16] M. Costabel, E. Stephan, A direct boundary integral equation method for transmission problems, J. Math. Anal. Appl. 106 (1985) 367-413. [17] S. L. Cotter, G. Roberts, A. M. Stuart, D. White, MCMC methods for functions: modifying old algorithms to make them faster, Statistical Science (2013), 424-446. [18] P. Dellaportas, J.J. Forster, I. Ntzoufras, On Bayesian model and variable selection using MCMC, Statistics and Computing 12 (2002) 27-36. [19] T.G. Dimiduk, V.N. Manoharan, Bayesian approach to analyzing holograms of colloidal particles, Optics Express 24 (2016) 24045-24060. [20] N. Dominguez, V. Gibiat, Non-destructive imaging using the time domain topological energy method, Ultrasonics 50 (2010), pp. 367-372. [21] V. Dom��nguez, S. Lu, F.J. Sayas, A Nystrom avored Calder�on calculus of order three for two dimensional waves, time-harmonic and transient. Comput. Math. Appl. 67 (2014), 217-236. [22] V. Dom��nguez, S. Lu, F.J. Sayas, A fully discrete Calderon calculus for two dimensional time harmonic waves. Int. J. Numer. Anal. Model. 11 (2014), 332-345. [23] O. Dorn, D. Lesselier, Level set methods for inverse scattering, Inverse Probl. 22 (2006) R67-R131. [24] M.M. Dunlop, M.A. Iglesias, A.M. Stuart, Hierarchical Bayesian level set inversion, Statistics and Computing, 27 (2017), 1555-1584. [25] M.M. Dunlop, Analysis and computation for Bayesian inverse problems, PhD Thesis, Warwick, 2016. [26] G.R. Feijoo, A new method in inverse scattering based on the topological derivative, Inverse Probl. 20 (2004) 1819-1840. [27] G.R. Feijoo, A.A. Oberai, P.M. Pinsky, An application of shape optimization in the solution of inverse acoustic scattering problems, Inverse Problems 20 (2004) 199-228. [28] R. Fletcher, Modi�ed Marquardt subroutine for non-linear least squares, Tech. Rep. 197213, 1971. [29] D. Foreman-Mackey, D.W. Hogg, D. Lang, J. Goodman, emcee: The MCMC Hammer, Publications of the Astronomical Society of the Paci�c 125(925), 2013. [30] H.W. Gao, K.I. Mishra, A. Winters, S. Wolin, D.G. Grier, Flexible wide-�eld high-resolution scanning camera for continuous-wave acoustic holography, Rev. Sci. Instrum. 89 (2018) 114901. [31] A. Gelman, D. B. Rubin, Inference from iterative simulation using multiple sequences, Statistical Science 7 (1992), 457{472. [32] J. Goodman, J. Weare, Ensemble samplers with a�ne invariance, Communications in Applied Mathematics and Computational Science 5 (2010), 65-80. [33] B. Guzina, M. Bonnet, Small-inclusion asymptotic of mis�t functionals for inverse problems in acoustics, Inverse Problems, 22 (2006) 1761-1785. [34] B. Guzina, F. Pourhamadian, Why the high-frequency inverse scattering by topological sensitivity may work, Proc. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci. 471 (2015) 2179. [35] L. He, C.Y. Kao, S. Osher, Incorporating topological derivatives into shape derivatives based level set methods, J. Comp. Phys. 225 (2007) 891-909. [36] F. Hettlich, Fr�echet derivatives in inverse obstacle scattering, Inverse Problems 11 (1995), 371-382. [37] M. Hintermuller, A. Laurain, Electrical impedance tomography: from topology to shape, Control and Cibernetics 37 (2008) 913-933. [38] T. Hohage, C. Schormann, A Newton-type method for a transmission problem in inverse scattering, Inverse Problems 14 (1998) 1207-1227. [39] K. Ito, B. Jin, J. Zou, A direct sampling method for inverse electromagnetic medium scattering, Inverse Problems 29 (2013) 095018. [40] J. Kaipio, E. Somersalo, Statistical and computational inverse problems, Vol. 160. Springer, (2006). [41] J.B. Keller, D. Givoli, Exact non-reecting boundary conditions, J. Comp. Phys. 82 (1989) 172-192. [42] A. Kirsch, The MUSIC-algorithm and the factorization method in inverse scattering theory for inhomogeneous media, Inverse problems 18 (2002) 1025-1040. [43] A. Klein, S. Falkner, S. Bartels, P. Hennig, F. Hutter, Fast Bayesian optimization of machine learning hyperparameters on large datasets, Proc. 20th Int. Conf. on Arti�cial Intelligence and Statistics (AISTATS) 2017, Fort Lauderdale, Florida, USA. JMLR: W&CP vol. 54, 2017. [44] R.E. Kleinman, P. Martin, On single integral equations for the transmission problem of acoustics, SIAM J. Appl. Math. 48 (1988) 307-325. [45] R. Kress, G.F. Roach, Transmission problems for the Helmholtz equation J. Math. Phys. 19 (1978) 1433-1437. [46] A. Lapidoth, A Foundation in Digital Communication, Cambridge University Press, 2009. [47] S.H. Lee, Y. Roichman, G.R. Yi, S.H. Kim, S.M. Yang, A. Blaaderen, P. van Oostrum, D.G. Grier, Characterizing and tracking single colloidal particles with video holographic microscopy, Optics Express 15 (2007) 18275-18282. [48] P. Li, Y. Wang, Z. Wang, Y. Zhao, Inverse obstacle scattering for elastic waves, Inverse Problems 32 (2016) 115018 [49] A. Litman, L. Crocco, Testing inversion algorithms against experimental data: 3D targets, Inverse Problems 25 (2009) 020201. [50] A. Palafox, M.A. Capistr�an, J.A. Christen, E�ective parameter dimension via Bayesian model selection in the inverse acoustic scattering problem, Mathematical Problems in Engineering 2014 (2014) 427203. [51] A. Palafox, Uncertainty quanti�cation on the inverse scattering problem, PhD Thesis, CIMAT, 2016. [52] A. Palafox, M.A. Capistr�an, J.A. Christen, Point cloud-based scatterer approximation and a�ne invariant sampling in the inverse scattering problem, Math. Meth. Appl. Sci. 40 (2017) 3393-3403. [53] W. Penny, J. Mattout, N Trujillo-Barreto, Bayesian model selection and averaging, Statistical Parametric Mapping: The analysis of functional brain images. London: Elsevier (2006). [54] N. Petra, J. Martin, G. Stadler, O. Ghattas, A computational framework for in�nite-dimensional Bayesian inverse problems, Part II: Stochastic Newton MCMC with application to ice sheet ow inverse problems, SIAM Journal on Scienti�c Computing 36 (2014), A1525-A1555. [55] Y. Rathi, N. Vaswani, A. Tannenbaum, A. Yezzi, Tracking deforming objects using particle �ltering for geometric active contours, IEEE T. Pattern Anal. 29 (2007), 1470-1475. [56] A. Tarantola, Inverse Problem Theory and Methods for Model Parameter Estimation, SIAM, Philadelphia, PA, (2005). [57] J. Sokolowski, A. Zochowski, On the topological derivative in shape optimization, SIAM Journal on Control and Optimization 37 (1999), 1251-1272. [58] H. Zhu, S. Li, S. Fomel, G. Stadler, O. Ghattas, A Bayesian approach to estimate uncertainty for full-waveform inversion using a priori information from depth migration, Geophysics 81 (2016) R307-R323. [59] T. Vincent, Introduction to Holography, CRC Press, Boca Raton, FL, 2012.