ANZIAM J. 46(E) ppC588--C622, 2005.

Inverse problems: A pragmatist's approach to the recovery of information from indirect measurements

R. S. (Bob) Anderssen

(Received 17 January 2005, revised 2 June 2005)

Abstract

Within the class of inverse problems, it is the subclass of indirect measurement problems that characterize the nature of inverse problems that arise in applications. With very few exceptions, measurements only record some indirect aspect of the phenomenon of interest (for example, X-rays and tomographic images in medical applications; telescope images in astronomy; stereological assessment of biological structure and processes; signatures in geophysical prospecting). Even when the direct information is measured such as weight or length, it is measured as a correlation against a standard and this correlation can be quite indirect, such as the measurement of weight by the extension (compression) of a spring.
The recovery of information about a phenomenon from indirect measurements is a piecemeal process. Any class of indirect measurements can only recover certain information about the phenomenon. In order to formulate realistic mathematical models that relate the indirect measurements to the specific information from the phenomenon that is to be recovered, there is a need to invoke simplifying assumptions (for example, radial or axial symmetry). The required information about the phenomenon is often only vaguely contained in the available indirect measurements, and this will be reflected in the nature of the mathematical equations which model the relationship between the indirect measurements and the phenomenon of interest.
All these aspects influence how the recovery of the information can be performed. The choice of methodology is not limited. The challenge is to perform the recovery in a way that correctly reflects the underlying nature of the problem context. It is not a matter of blindly applying some form of quadratic regularization for which algorithms and packages are readily available. Though such tools are useful for initial exploratory analysis, the crucial characterization of the information to be recovered is hidden in the mathematical model that relates the indirect measurements to the phenomenon within the problem context.
When recovering information from indirect measurements, the question that focuses the problem solving comes from the need for decision making to have answers to specific matters. The data available for the associated decision support will be indirect measurements of the phenomenon under investigation. As a consequence of the applications context, the recovery of information of the associated inverse problem will be constrained by practical challenges including:
  1. In a given situation, how does one decide on the indirect measurements to be performed?
  2. How are some practical people able to solve indirect measurement problems without having to perform an explicit regularization?
  3. Is there any advantage in combining different indirect measurements of the same phenomenon?
  4. What are the alternatives, when there is only a (very) limited amount of data?
  5. How does one proceed when a mathematical model is not available or is too complex to formulate?
We examine, in terms of practical problems, how such challenges can be accommodated, as well as explore the wide range of mathematical matters and considerations that arise when solving inverse problems.

Download to your computer

Authors

R. S. (Bob) Anderssen
CSIRO Mathematical and Information Sciences, GPO Box 664, Canberra, ACT 2601, Australia. mailto:Bob.Anderssen@csiro.au

Published July 9, 2005. ISSN 1446-8735

References

  1. D. L. Altshuler, R. Dudley and C. P. Ellington, Aerodynamic forces of revolving hummingbird wings and wing models, J. Zoology 264 (2004), 327--332. http://dx.doi.org/10.1017/S0952836904005813
  2. M. C. Anderson, Studies of the woodland climate. I. The photographic computation of light conditions, J. Ecol. 52 (1964), 27--41.
  3. M. C. Anderson, Radiation and crop structure, in Plant Photosynthetic Production, Manual of Methods (Eds J. Catsky and P. G. Jarvis), Junk, The Hague, 1971, p. 412--466.
  4. R. S. Anderssen, On the use of linear functionals for Abel-type integral equations in applications, in Applications and Numerical Solution of Integral Equations (Eds. R. S. Anderssen, F. R. de Hoog and M. A. Lukas), Sijthoff and Noordhoff, The Netherlands, 1980, pp. 195--221.
  5. R. S. Anderssen, The linear functional strategy for improperly posed problems, in Inverse Problems (Eds. J. R. Cannon and U. Hornung; Proceedings of the Oberwolfach Conference, May, 1986), Birkhauser Verlag, Basel, 1986, pp. 11--30.
  6. R. S. Anderssen, The pragmatics of solving industrial (real-world) inverse problems with exemplification based on the molecular weight distribution problem, Inverse Problems 15 (1999), R1-R40.
  7. R. S. Anderssen and A. R. Davies, Simple moving-average formulae for the direct recovery of the relaxation spectrum, J. Rheol. 45 (2001), 1--27. http://dx.doi.org/10.1122/1.1332787
  8. R. S. Anderssen and F. R. de Hoog, On the method of Chan and Lu for Abel's integral equation, J. Phys. A: Math. Gen. 14 (1981), 3117--3121.
  9. R. S. Anderssen and F. R. de Hoog, Finite difference methods for the numerical differentiation of non-exact data, Computing 33 (1984), 259--267.
  10. R. S. Anderssen and F. R. de Hoog, Abel integral equations, in Numerical Solution of Integral Equations (Ed. M. A. Golberg), Plenum Press, New York, 1990, pp. 373--410.
  11. R. S. Anderssen, F. R. de Hoog and M. Hegland, A stable finite difference ansatz for higher order differentiation of non-exact data, Bull. Austral. Math. Soc. 58 (1998), 223--232.
  12. R. S. Anderssen and M. Hegland, For numerical differentiation, dimensionality can be a blessing! Math. Comp. 68 (1999), 1121--1141.
  13. R. S. Anderssen, F. R. de Hoog and M. A. Lukas, The Application and Numerical Solution of Integral Equations, Sijthoff & Noordhoff, Alphen aan den Rijn, The Netherlands, 1980.
  14. R. S. Anderssen and J. R. Jackett, Linear functionals of the foliage angle density, J. Aust. Math. Soc., Series B 25 (1984), 431--442.
  15. R. S. Anderssen and D. R. Jackett, Use of the functional strategy for assessing the status of plant canopies, Chapter 7 in {Industrial Numerical Analysis} (Editors: S. McKee and C. M. Elliott), Clarendon Press, Oxford, 1986.
  16. R. S. Anderssen, D. R. Jackett and D. L. B. Jupp, Linear functionals of the foliage angle distribution as tools to study the structure of plant canopies, Aust. J. Bot. 32 (1984), 147--156.
  17. R. S. Anderssen, S. A. Husain and R. J. Loy, The Kohlrausch function: properties and applications, ANZIAM J. 45(E) (2004), C800--C816. http://anziamj.austms.org.au/V45/CTAC2003/Ande
  18. R. S. Anderssen and R. J. Loy, Completely monotone fading memory relaxation modulii, Bull. Austral. Math. Soc. 65 (2002), 449--460.
  19. R. S. Anderssen and R. J. Loy, Rheological implications of completely monotone fading memory, J. Rheol. 46 (2002), 1459--1472. http://dx.doi.org/10.1122/1.1514203
  20. R. S. Anderssen, E. Carter, B. J. Osborne and I. J. Wesley, Joint inversion of multi-modal spectroscopic data of wheat flours, Applied Spectroscopy (in press)
  21. R. S. Anderssen, B. J. Osborne and I. J. Wesley, The application of localisation to near infrared calibration and prediction through partial least squares regression, J. Near Infrared Spectrosc. 11 (2003), 39--48.
  22. D. L. Bailey, B. F. Hutton and B. J. Walker, Improved SPECT using simultaneous emission and transmission tomography, J. Nuclear Med. 28 (1987), 844--851.
  23. M. Burger and S. Osher, Convergence rates of convex variational regularization, Inverse Problems 20 (2004), 1411--1421.
  24. L. Boltzmann, Zur Theorie der elastischen Nachwirkung, Ann. Phys. Chem. 7 (1876), 624--657.
  25. J. A. Cochran, Applied Mathematics Wadsworth, Belmont, California, 1982.
  26. A. R. Davies and R. S. Anderssen, Sampling localization in determining the relaxation spectrum, J. Non-Newtonian Fluid Mech. 73 (1997), 163--179.
  27. J. des Cloizeaux, Double reptation vs single reptation in polymer melts, Europhys Lett 5 (1988), 437--442; 6 (1988), 475.
  28. M. Doi and S. F. Edwards, Dynamics of concentrated polymer systems. Part 1. Brownian motion in the equilibrium state, J. Chem. Soc., Faraday Trans. 274 (1978), 1789--1801.
  29. D. Donoho, Nonlinear solution of linear inverse problems by wavelet-vaguelette decomposition, Appl. Comp. Harmonic Anal. 2 (1995), 101--126. http://dx.doi.org/10.1006/acha.1995.1008
  30. H. W. Engl, M. Hanke and A. Neubauer, Regularization of Inverse Problems, Kluwer Academic Publishers, Boston, 1996.
  31. G. M. L. Gladwell, Inverse Problems in Vibration, Martinus Noordhoff (Kluwer), The Netherlands, 1986.
  32. M. Golberg, A method of adjoints for solving some ill-posed inverse problems, Appl. Math. Comput. 5 (1979), 123--130.
  33. M. A. Golberg, Numerical Solution of Integral Equations, Plenum Press, New York, 1990.
  34. C. W. Groetsch, Inverse Problems in the Mathematical Sciences, Vieweg, Wiesbaden, 1993.
  35. S. A. Husain and R. S. Anderssen, Modelling the relaxation modulus of linear viscoelasticity using Kohlrausch functions, J. Non-Newtonian Fluid Mech. 125 (2005), 159--170. http://dx.doi.org/10.1016/j.jnnfm.2004.11.002
  36. A. J. Jakeman and R. S. Anderssen, Abel type integral equations in stereology. I. General discussion, J. Microscopy 105 (1975), 121--133.
  37. M. G. Kendall and P. A. P. Moran, Geometric Probability, Griffins Statistical Monographs, No. 10, C. Griffin, London, 1963.
  38. I. Koch and R. S. Anderssen, A direct surface smoothing procedure for Fourier image reconstruction in radio-physics, Astron. Astrophys. 183 (1987), 170--176.
  39. G. Landl and R. S. Anderssen, Non-negative differentially constrained entropy-like regularization, Inverse Problems 12 (1996), 35--53.
  40. H. Martens and T. Naes, Multivariate Calibration, John Wiley & Sons, Chichester, UK, 1991.
  41. C. R. Martin, R. Rousser and D. L. Barbec, Development of a single-kernel characterization system, Trans. ASAE 36 (1993), 1399--1404.
  42. P. Mathe and S. V. Pereverzev, Direct estimation of linear functionals from indirect noisy observations, J. Complexity 18 (2002), 500--516. http://dx.doi.org/10.1006/jcom.2001.0614
  43. A. McNabb and G. C. Wake, Heat conduction and finite measures for transition times between steady states, IMA J. Appl. Maths. 47 (1991), 193--260.
  44. A. McNabb, G. C. Wake, R. D. Lambourne and R. S. Anderssen, The theoretical derivation of rules-of-thumb for freezing times, Inverse Problems 7 (1991), 633--642.
  45. D. W. Mead, Determination of molecular weight distributions of linear flexible polymers from linear viscoelastic material functions, J. Rheol. 38 (1994), 1797--1827. http://dx.doi.org/10.1122/1.550527
  46. J. B. Miller, An integral equation from Phytology, J. Aust. Math. Soc. 4 (1963), 397--402.
  47. P. A. P. Moran, The probabilistic basis of stereology, Special Supplement Adv. Appl. Probability 4 (1972), 69--91.
  48. V. A. Morozov, Regularization Methods for Ill-Posed Problems, CRC Press, 1993.
  49. D. Murio, The Mollification Method and the Numerical Solution of Ill-Posed Problems, John Wiley & Sons, 1993.
  50. T. Naes, T. Isaksson, T. Fearn and T. Davies, A User-Friendly Guide to Multivariate Calibration and Classification, NIR Publications, Chichester, UK, 2002.
  51. B. G. Osborne and R. S. Anderssen, Single-kernel characterization principles and applications, Cereal Chem. 80 (2003), 613--622. http://dx.doi.org/10.1098/rstb.2003.1352
  52. W. Schmeidler, Integralgleichungen mit Anwendungen in Physik und Technik, Geest & Portig, Leipzig, 1950.
  53. D. H. Simmonds, Wheat and Wheat Quality in Australia, CSIRO Publishing, 1989.
  54. J. A. Smith, R. E. Oliver and J. K. Berry, A comparison of two photographic techniques for estimating foliage angle distribution, Aust. J. Bot. 25 (1977), 545--553.
  55. G. R. Spedding, Comparing fluid mechanics models with experimental data, Phil. Trans. Proc. Roy. Soc. B 358 (2003), 1567--1576.
  56. W. Thimm, C. Friedrich, M. Marth and J. Honerkamp, An analytic relation between relaxation time spectrum and molecular weight distribution, J. Rheol. 43 (1999), 1663--1672. http://dx.doi.org/10.1122/1.551066
  57. C. Tsenoglou, Molecular weight polydispersity effects on the viscoelasticity of entangled polymers, Macromolecules 24 (1991), 1762--1767.
  58. C. Vogel, Computational Methods for Inverse Problems, SIAM, Philadelphia, 2002
  59. G. Wahba, Spline Models for Observational Data, SIAM, Philadelphia, 1990.
  60. J. Warren Wilson, Inclined point quadrats, with Appendix by J. E. Reeve, The New Phytologist 59 (1960), 1--8.
  61. C. Zimmer, By a whisker, harbour seals catch their prey, Science 293 (2001), 29--31. http://dx.doi.org/10.1126/science.293.5527.29b