pith. sign in

arxiv: 2607.02180 · v1 · pith:DESI7YXOnew · submitted 2026-07-02 · 🧮 math.NA · cs.NA

A highly efficient iterative approach for inverse acoustic obstacle scattering problems in three dimensions

Zhiyong Cheng , Heping Dong , Lu Zhao This is my paper

Pith reviewed 2026-07-03 07:47 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords inverse acoustic scatteringobstacle reconstructioniterative methodhomothetic surfaceboundary integral equationFréchet derivativefar-field datathree dimensions
0
0 comments X

The pith

An iterative method reconstructs three-dimensional acoustic obstacles from scattered or far-field data using boundary integrals on homothetic surfaces that avoid all singularities.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes an iterative reconstruction technique for inverse acoustic obstacle scattering in three dimensions that defines all boundary integrals on a homothetic surface instead of the unknown obstacle boundary. This choice lets the method prove that the Fréchet derivative of the far-field map is injective with dense range, which guarantees solvability of the linearized data equation at each step. The same homothetic surface is shown to generate a scattered field that can approximate the exact scattered field to any desired accuracy. A sympathetic reader would care because conventional boundary-integral approaches in three dimensions must handle singular kernels directly on the obstacle surface, and the new construction removes that difficulty while still recovering the shape from either phased or phaseless data.

Core claim

The central claim is that an iterative scheme based on boundary integral equations defined on a homothetic surface reconstructs the obstacle from scattered-field or phased/phaseless far-field data in three dimensions; the scheme completely avoids singularities because the scattered field produced by the homothetic surface can arbitrarily approximate the exact scattered field, while the injectivity and dense-range property of the associated Fréchet derivative ensure that the linearized data equation remains solvable at every iteration.

What carries the argument

The homothetic surface, on which boundary integrals are defined so that the generated scattered field approximates the exact one and the Fréchet derivative of the far-field operator acquires injectivity plus dense range.

If this is right

  • The same iterative scheme applies equally to phased and phaseless far-field data.
  • No explicit treatment of singular kernels on the obstacle surface is required at any stage.
  • The dense-range property of the Fréchet derivative guarantees that the linearized equation can be solved at each iteration.
  • Numerical tests confirm that the method remains stable under moderate noise levels.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The construction may extend directly to other inverse scattering problems whose forward maps admit similar Fréchet-derivative properties.
  • Adaptive selection of the homothetic ratio could be tested to improve convergence speed for elongated obstacles.
  • The avoidance of singularities suggests the method could be combined with fast multipole or fast Fourier techniques for large-scale three-dimensional reconstructions.

Load-bearing premise

The homothetic surface must supply a scattered-field approximation accurate enough for the iteration to converge to the true obstacle boundary.

What would settle it

A concrete numerical experiment on a known smooth obstacle in which the iteration, started from the homothetic surface, fails to recover the correct shape even though the theoretical approximation property is claimed to hold.

Figures

Figures reproduced from arXiv: 2607.02180 by Heping Dong, Lu Zhao, Zhiyong Cheng.

Figure 1
Figure 1. Figure 1: True shape of 3D plots with front, side and top views. In all figures, the shaded regions represent the projections of the reconstructions, whereas the colored dashed curves denote the true obstacle boundaries in the corresponding projection directions. This visualization allows a direct comparison between the exact geometry and the recovered shape. We begin with two standard test obstacles: a pinched ball… view at source ↗
Figure 2
Figure 2. Figure 2: Reconstructions of a pinched ball-shaped obstacle with single incident wave. Subfigures (a)-(c) show results from phased scattered field, phased far-field, and phaseless far-field data, respectively, each with 1%, 5% and 10% noise. The stopping parameters ϵ = (0.009, 0.023, 0.045) for (a), ϵ = (0.018, 0.032, 0.050) for (b), and ϵ = (0.004, 0.015, 0.030) for (c). settings are kept unchanged except the numbe… view at source ↗
Figure 3
Figure 3. Figure 3: Reconstructions of a cushion-shaped obstacle with single incident wave. Subfigures (a)-(c) show results from phased scattered field, phased far-field, and phaseless far-field data, respectively, each with 1%, 5% and 10% noise. The stopping parameters ϵ = (0.009, 0.023, 0.045) for (a), ϵ = (0.018, 0.032, 0.050) for (b), and ϵ = (0.004, 0.015, 0.030) for (c). and third rows illustrate results for different w… view at source ↗
Figure 4
Figure 4. Figure 4: Reconstructions of a cushion shaped obstacle with multiple incident waves under 5% noise. Subfigures (a), (b), and (c) correspond to reconstructions with two incident waves using phased scattered field, phased far field, and phaseless far field data, respectively. Subfigures (d), (e), and (f) correspond to reconstructions with four incident waves under the same three data types. The initial guesses, wavenu… view at source ↗
Figure 5
Figure 5. Figure 5: Reconstructions of a cushion-shaped obstacle from phaseless far-field data with multiple point sources under 5% noise. Subfigures (a)–(c) show the results ob￾tained with one, two, and four point sources, respectively. In each subfigure, the four panels show the three-dimensional reconstruction and its projection views along the directions (1, 0, 0)⊤, (0, 1, 0)⊤, and (0, 0, 1)⊤. In all cases, c (0) = (0.1, … view at source ↗
Figure 6
Figure 6. Figure 6: Reconstruction of a pinched ball-shaped obstacle from phased scattered field data generated by two incident waves in the directions (0, 0, 1)⊤ and (0, 0, −1)⊤ with 5% noise, ϵ = 0.023. To examine the performance of the proposed approach under incomplete angular information, we consider the reconstruction of the cushion-shaped obstacle under the Dirichlet boundary condition using phaseless far-field data me… view at source ↗
Figure 7
Figure 7. Figure 7: Reconstruction of a cushion-shaped obstacle from phased far-field data generated by two incident waves (1, 0, 0)⊤, (−1, 0, 0)⊤ with 5% noise, ϵ = 0.032. by the radial function r(θ, ϕ) = 0.6  1 + 0.3 sin(7θ) cos ϕ + 0.2 sin2 (3θ) sin(2ϕ) + 0.1 cos θ  . The surface exhibits strong angular oscillations in both θ and ϕ, resulting in multiple localized peaks, dents, and nonconvex regions, making this example … view at source ↗
Figure 8
Figure 8. Figure 8: Reconstruction of a pinched ball-shaped obstacle from phaseless far-field data generated by two point sources located at (4, 0, 0)⊤ and (−4, 0, 0)⊤ with 5% noise, ϵ = 0.015. We also note that Example 7 involves substantially higher computational complexity due to the increased number of quadrature points, observation directions, and the larger truncation parameter Mmax, which together result in a significa… view at source ↗
Figure 9
Figure 9. Figure 9: Reconstructions of a cushion-shaped obstacle from phaseless far-field data under different boundary conditions with 1% noise. The incident field is generated by point sources located at (4, 0, 0)⊤ and (−4, 0, 0)⊤, and the initial guess is a sphere with c (0) = (0, −0.1, 0.1)⊤ and r (0) = 0.6. (a) Initial surface (b) M = 1 (c) M = 3 (d) M = 5 [PITH_FULL_IMAGE:figures/full_fig_p025_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Reconstruction of a cushion-shaped obstacle from phased scattered field data with 1% noise. κ = 4, d = (±1, 0, 0)⊤, R = 5, c (0) = (0.2, −0.6, 0.3)⊤, r (0) = 0.3, ϵ = 0.009. Subfigures (a)–(d) show the initial surface and the progressive reconstructions for truncation numbers M = 1, 3, 5, respectively [PITH_FULL_IMAGE:figures/full_fig_p025_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Reconstruction of a bean-like surface obstacle under the Dirichlet boundary condition from phaseless far-field pattern with 1% noise. (a) shows the true shape; (b) shows the reconstructions from two point sources located at (4, 0, 0)⊤ and (−4, 0, 0)⊤; (c) shows the reconstructions from four point sources located at (±4, 0, 0)⊤ and (0, ±4, 0)⊤. The initial guess is a sphere with c (0) = (0.1, −0.5, 0.1)⊤ a… view at source ↗
Figure 12
Figure 12. Figure 12: Reconstruction of a cushion-shaped obstacle under the Dirichlet bound￾ary condition from limited-aperture phaseless far-field data generated by two point sources located at (4, 0, 0)⊤ and (−4, 0, 0)⊤ with 1% noise. c (0) = (0.2, −0.3, −0.1)⊤, r (0) = 0.4, κ = 4.5, ϵ = (0.005, 0.006, 0.008). The corresponding data equations on the observation sphere are K1g1,t + K2g2,t = u sc t,δ|ΓBR , t = 1, . . . , 6, wi… view at source ↗
Figure 13
Figure 13. Figure 13: Reconstruction of a complex obstacle under the Dirichlet boundary condition from phaseless far-field data generated by four point sources located at (0, 0, ±4)⊤ and (±4, 0, 0)⊤ with different noise levels. c (0) = (0.1, −0.5, 0.1)⊤, r (0) = 0.3, κ = 5. homothetic surface can arbitrarily approximate the exact one. Moreover, we established that the associated Fr´echet derivative is injective and has dense r… view at source ↗
Figure 14
Figure 14. Figure 14: Reconstruction of a complex obstacle under the Dirichlet boundary condition from phaseless far-field data generated by four point sources located at (0, 0, ±4)⊤ and (±4, 0, 0)⊤ with 1% noise. c (0) = (0.1, −0.5, 0.1)⊤, r (0) = 0.3, κ = 5. Subfigures (a)–(g) show the initial surface and the progressive reconstructions obtained as the truncation number M increases. Subfigure (h) reports the relative error w… view at source ↗
Figure 15
Figure 15. Figure 15: Reconstruction of two sound-soft obstacles from 1% noisy scattered￾field data. We set N˜ = 2 with initial centers c (0) 1 = (−1.3, 0.4, 0.2)⊤ and c (0) 2 = (1.3, −0.2, −0.1)⊤. The parameters are κ = 5, r (0) = 0.4, ς = 0.9, α = λ = 10−8 , ρ = 0.05, Mmax = 6, and ϵ = 0.0065 [PITH_FULL_IMAGE:figures/full_fig_p029_15.png] view at source ↗
read the original abstract

This paper concerns a three-dimensional inverse acoustic obstacle scattering problem from scattered field or phased/phaseless far-field data. Based on the boundary integral defined on a homothetic surface, we propose a highly efficient iterative approach for obstacle reconstruction that completely avoids dealing with any singularity. Here, the injectivity and dense-range property of the Fr\'echet derivative have been proved to ensure the solvability of the linearized equivalent data equation. We also prove that the scattered field generated by the homothetic surface can arbitrarily approximate the exact one. Numerical experiments are presented to verify the superiority and robustness of the proposed approach.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper proposes an iterative method for reconstructing 3D acoustic obstacles from scattered-field or phased/phaseless far-field data. The method defines boundary integrals on a homothetic surface to avoid singularities entirely. It proves injectivity and dense range of the Fréchet derivative of the forward map to guarantee solvability of each linearized data equation, proves that the scattered field generated by the homothetic surface can arbitrarily approximate the exact scattered field, and reports numerical experiments showing superiority and robustness.

Significance. If the approximation property can be equipped with quantitative rates and stability estimates, the approach would offer a genuinely singularity-free, provably solvable iteration for a core inverse-scattering problem; the combination of functional-analytic guarantees (injectivity/dense range) with a modeling choice that sidesteps singular integrals is a clear technical strength.

major comments (2)
  1. [Abstract] Abstract (paragraph on the homothetic surface): the claim that the scattered field generated by the homothetic surface “can arbitrarily approximate the exact one” is stated without an explicit rate or stability estimate in any Sobolev norm on the far-field sphere or on the boundary. Because the iteration relies on this approximation remaining sufficiently accurate for the update direction to remain valid at each step, the absence of a quantitative bound linking the homothety ratio to the residual of the data equation is load-bearing for convergence of the overall scheme.
  2. [Abstract] Abstract (paragraph on the Fréchet derivative): while injectivity and dense range are asserted to ensure solvability of the linearized equation, the manuscript does not indicate whether these properties are uniform with respect to the homothetic surface or degrade as the current guess moves away from the true obstacle; without such uniformity the solvability result does not automatically transfer to the practical iteration.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and the constructive major comments. We address each point below, indicating where revisions will be made to improve clarity.

read point-by-point responses

  1. Referee: [Abstract] Abstract (paragraph on the homothetic surface): the claim that the scattered field generated by the homothetic surface “can arbitrarily approximate the exact one” is stated without an explicit rate or stability estimate in any Sobolev norm on the far-field sphere or on the boundary. Because the iteration relies on this approximation remaining sufficiently accurate for the update direction to remain valid at each step, the absence of a quantitative bound linking the homothety ratio to the residual of the data equation is load-bearing for convergence of the overall scheme.

    Authors: The manuscript establishes that the difference between the scattered field on the homothetic surface and the exact scattered field can be made arbitrarily small by taking the homothety ratio sufficiently close to one; this is a qualitative density-type result rather than a quantitative rate. In the iteration the homothety ratio is chosen once and for all close enough that the modeling error is smaller than the current data residual, which is justified by the arbitrary-approximation statement. We agree that an explicit rate would be desirable and will add a short clarifying paragraph in the revised manuscript noting that the present proof is qualitative and that quantitative estimates remain an open question for future work. The numerical section already shows robust convergence for the ratios used in practice. revision: partial

  2. Referee: [Abstract] Abstract (paragraph on the Fréchet derivative): while injectivity and dense range are asserted to ensure solvability of the linearized equation, the manuscript does not indicate whether these properties are uniform with respect to the homothetic surface or degrade as the current guess moves away from the true obstacle; without such uniformity the solvability result does not automatically transfer to the practical iteration.

    Authors: The injectivity and dense-range proofs are given for the Fréchet derivative of the forward map defined on any fixed homothetic surface lying sufficiently close to the unknown obstacle. At each step of the iteration the current surface is regarded as fixed, so the solvability result applies directly to the linearized equation at that iterate. The manuscript does not claim uniformity over the entire admissible set of surfaces; the local nature of the argument is implicit in the statement that the homothetic surface approximates the true obstacle. We will insert one clarifying sentence in the revised version to make the per-iterate applicability explicit. revision: partial

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The derivation chain rests on standard boundary-integral scattering theory plus two independent proofs (injectivity/dense-range of the Fréchet derivative and the approximation property of the homothetic-surface field). Neither proof is shown to reduce by construction to a fitted parameter, a self-citation chain, or a renaming of an input; the homothetic modeling choice is an explicit ansatz whose approximation claim is asserted to be proved rather than assumed. The method therefore remains self-contained against external benchmarks and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper rests on standard acoustic scattering assumptions; the homothetic surface is a modeling device rather than a new physical entity.

axioms (2)
  • domain assumption The forward scattering problem is governed by the Helmholtz equation with sound-hard or impedance boundary conditions on the obstacle.
    Standard modeling assumption for acoustic obstacle scattering invoked throughout the inverse-problem setup.
  • domain assumption Far-field or phased/phaseless scattered-field data are available and contain sufficient information for shape recovery.
    Implicit premise of the inverse problem formulation.

pith-pipeline@v0.9.1-grok · 5627 in / 1268 out tokens · 27116 ms · 2026-07-03T07:47:43.471841+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

31 extracted references

  1. [1]

    Ammari,An introduction to mathematics of emerging biomedical imaging, Math´ ematiques & Applications (Berlin), 62, Springer, Berlin, 2008

    H. Ammari,An introduction to mathematics of emerging biomedical imaging, Math´ ematiques & Applications (Berlin), 62, Springer, Berlin, 2008

    2008

  2. [2]

    Ammari, G

    H. Ammari, G. Bao and J. L. Fleming, An inverse source problem for Maxwell’s equations in magnetoencephalog- raphy, SIAM J. Appl. Math. 62 (2002) 1369–1382

    2002

  3. [3]

    G. Bao, P. Li, Inverse medium scattering problems for electromagnetic waves, SIAM J. Appl. Math. 65 (2005) 2049–2066

    2005

  4. [4]

    Bonnet and F

    M. Bonnet and F. Cakoni, Analysis of topological derivative as a tool for qualitative identification, Inverse Probl. 35 (2019) 104007

    2019

  5. [5]

    Borden, Mathematical problems in radar inverse scattering, Inverse Probl

    B. Borden, Mathematical problems in radar inverse scattering, Inverse Probl. 18 (2002) R1–R28

    2002

  6. [6]

    Borges, J

    C. Borges, J. Lai, Inverse scattering reconstruction of a three-dimensional sound-soft axis-symmetric impenetrable object, Inverse Probl. 36 (2020) 105005

    2020

  7. [7]

    Carpio, T

    A. Carpio, T. Dimiduk, F. Le Lou¨ er, M. Rap´ un, When topological derivatives met regularized Gauss-Newton iterations in holographic 3D imaging, J. Comput. Phys. 388 (2019) 224–251

    2019

  8. [8]

    Chang, Y

    Y. Chang, Y. Guo, H. Liu, D. Zhang, Recovering source location, polarization, and shape of obstacle from elastic scattering data, J. Comput. Phys. 489 (2023) 112289

    2023

  9. [9]

    J. Chen, Z. Chen and G. Huang, Reverse time migration for extended obstacles: acoustic waves, Inverse Probl. 29 (2013) 085005

    2013

  10. [10]

    J. Chen, B. Jin, H. Liu, Solving inverse obstacle scattering problem with latent surface representations, Inverse Probl. 40 (2024) 065013

    2024

  11. [11]

    Colton, R

    D. Colton, R. Kress, Inverse Acoustic and Electromagnetic Scattering Theory, 4th ed., Springer, New York, 2019

    2019

  12. [12]

    Colton, J

    D. Colton, J. Coyle, P. Monk, Recent developments in inverse acoustic scattering theory, SIAM Rev. 42 (2000) 369–414

    2000

  13. [13]

    H. Dong, J. Lai, P. Li, Inverse obstacle scattering for elastic waves with phased or phaseless far-field data, SIAM J. Imaging Sci. 12 (2019) 809–838

    2019

  14. [14]

    H. Dong, J. Lai and P. Li, A spectral boundary integral method for the elastic obstacle scattering problem in three dimensions, J. Comput. Phys. 469 (2022) 111546

    2022

  15. [15]

    Farhat, R

    C. Farhat, R. Tezaur, R. Djellouli, On the solution of three-dimensional inverse obstacle acoustic scattering problems by a regularized Newton method, Inverse Probl. 18 (2002) 1229–1246

    2002

  16. [16]

    Ganesh, I

    M. Ganesh, I. Graham, A high-order algorithm for obstacle scattering in three dimensions, J. Comput. Phys. 198 (2004) 211–242

    2004

  17. [17]

    Hagemann, T

    F. Hagemann, T. Arens, T. Betcke, F. Hettlich, Solving inverse electromagnetic scattering problems via domain derivatives, Inverse Probl. 35 (2019) 084005

    2019

  18. [18]

    Hohage, Logarithmic convergence rates of the iteratively regularized Gauss-Newton method for an inverse potential and an inverse scattering problem, Inverse Probl

    T. Hohage, Logarithmic convergence rates of the iteratively regularized Gauss-Newton method for an inverse potential and an inverse scattering problem, Inverse Probl. 13 (1997) 1279–1299

    1997

  19. [19]

    Ivanyshyn, R

    O. Ivanyshyn, R. Kress, P. Serranho, Huygens’ principle and iterative methods in inverse obstacle scattering, Adv. Comput. Math. 33 (2010) 413–429

    2010

  20. [20]

    Ivanyshyn, R

    O. Ivanyshyn, R. Kress, Identification of sound-soft 3D obstacles from phaseless data, Inverse Probl. Imaging 4 (2010) 131–149

    2010

  21. [21]

    Kang and M

    S. Kang and M. Lambert, Structure analysis of direct sampling method in 3D electromagnetic inverse problem: near- and far-field configuration, Inverse Probl. 37 (2021) 075002

    2021

  22. [22]

    Kirsch, Characterization of the shape of a scattering obstacle using the spectral data of the far field operator, Inverse Probl

    A. Kirsch, Characterization of the shape of a scattering obstacle using the spectral data of the far field operator, Inverse Probl. 14 (1998) 1489–1512

    1998

  23. [23]

    Kress and A

    R. Kress and A. Zinn, On the numerical solution of the three-dimensional inverse obstacle scattering problem, J. Comput. Appl. Math. 42 (1992) 49–61

    1992

  24. [24]

    Kress, Newton’s method for inverse obstacle scattering meets the method of least squares, Inverse Probl

    R. Kress, Newton’s method for inverse obstacle scattering meets the method of least squares, Inverse Probl. 19 (2003) S91–S104

    2003

  25. [25]

    Le Lou¨ er and M.-L

    F. Le Lou¨ er and M.-L. Rap´ un, Detection of multiple impedance obstacles by non-iterative topological gradient based methods, J. Comput. Phys. 388 (2019) 534–560

    2019

  26. [26]

    Le Lou¨ er and M.-L

    F. Le Lou¨ er and M.-L. Rap´ un, A boundary integral formulation and a topological energy-based method for an inverse 3D multiple scattering problem with sound-soft, sound-hard, penetrable, and absorbing objects, Comput. Methods Appl. Math. 22 (2022) no. 4, 915–943. AN ITERATIVE APPROACH FOR INVERSE ACOUSTIC OBSTACLE SCATTERING 31

    2022

  27. [27]

    Li and X

    P. Li and X. Yuan, Inverse obstacle scattering for elastic waves in three dimensions, Inverse Probl. Imaging 13 (2019) 545–573

    2019

  28. [28]

    Liu and K

    H. Liu and K. Liu, Direct imaging of inhomogeneities in a 3D shallow ocean waveguide with an icecap, J. Comput. Phys. 498 (2024) 112694

    2024

  29. [29]

    Potthast, Point Sources and Multipoles in Inverse Scattering Theory, Chapman & Hall/CRC, Boca Raton, FL, 2001

    R. Potthast, Point Sources and Multipoles in Inverse Scattering Theory, Chapman & Hall/CRC, Boca Raton, FL, 2001

    2001

  30. [30]

    Serranho, A hybrid method for inverse scattering for a sound-soft obstacle in R3, Inverse Probl

    P. Serranho, A hybrid method for inverse scattering for a sound-soft obstacle in R3, Inverse Probl. Imaging 1 (2007) 691–712

    2007

  31. [31]

    K. Ito, B. Jin, J. Zou, A direct sampling method to an inverse medium scattering problem, Inverse Probl. 28 (2012) 025003. Jilin University, Qianjin Street 2699, Changchun, Jilin 130012, China Email address:chengzy22@mails.jlu.edu.cn Jilin University, Qianjin Street 2699, Changchun, Jilin 130012, China Email address:dhp@jlu.edu.cn Civil Aviation Universit...

    2012