Skip to main content

Anisotropic multiscale systems on bounded domains

Abstract

We establish a construction of hybrid multiscale systems on a bounded domain \({\Omega } \subset \mathbb {R}^{2}\) consisting of shearlets and boundary-adapted wavelets, which satisfy several properties advantageous for applications to imaging science and the numerical analysis of partial differential equations. More precisely, we construct hybrid shearlet-wavelet systems that form frames for the Sobolev spaces \(H^{s}({\Omega }),~s\in \mathbb {N} \cup \{0\}\) with controllable frame bounds and admit optimally sparse approximations for functions which are smooth apart from a curve-like discontinuity. Per construction, these systems allow to incorporate boundary conditions.

This is a preview of subscription content, access via your institution.

References

  1. 1.

    Bittner, K.: Biorthogonal spline wavelets on the interval. In: Wavelets and splines: Athens 2005, pp 93–104. Nashboro Press, Brentwood (2006)

  2. 2.

    Candès, E.J.: Ridgelets: Theory and applications. Stanford University, PhD thesis (1998)

    Google Scholar 

  3. 3.

    Candès, E.J., Demanet, L.: The curvelet representation of wave propagators is optimally sparse. Comm. Pure Appl. Math. 58(11), 1472–1528 (2005)

    MathSciNet  MATH  Article  Google Scholar 

  4. 4.

    Candès, E.J., Donoho, D.L.: New tight frames of curvelets and optimal representations of objects with piecewise C2 singularities. Comm. Pure Appl. Math. 57(2), 219–266 (2004)

    MathSciNet  MATH  Article  Google Scholar 

  5. 5.

    Canuto, C., Tabacco, A., Urban, K.: The wavelet element method. I. Construction and analysis. Appl. Comput. Harmon. Anal. 6(1), 1–52 (1999)

    MathSciNet  MATH  Article  Google Scholar 

  6. 6.

    Christensen, O.: An introduction to frames and Riesz bases. Applied and numerical harmonic analysis. Birkhäuser Boston, Inc, Boston (2003)

    MATH  Google Scholar 

  7. 7.

    Cohen, A., Dahmen, W., DeVore, R. : Multiscale decompositions on bounded domains. Trans. Amer. Math Soc. 352(8), 3651–3685 (2000)

    MathSciNet  MATH  Article  Google Scholar 

  8. 8.

    Cohen, A., Dahmen, W., DeVore, R.: Adaptive wavelet methods for elliptic operator equations: convergence rates. Math. Comp. 70(233), 27–75 (2001)

    MathSciNet  MATH  Article  Google Scholar 

  9. 9.

    Cohen, A., Daubechies, I., Vial, P.: Wavelets on the interval and fast wavelet transforms. Appl. Comput. Harmon. Anal. 1(1), 54–81 (1993)

    MathSciNet  MATH  Article  Google Scholar 

  10. 10.

    Dahlke, S., Fornasier, M., Raasch, T.: Adaptive frame methods for elliptic operator equations. Adv Comput. Math. 27(1), 27–63 (2007)

    MathSciNet  MATH  Article  Google Scholar 

  11. 11.

    Dahlke, S., Raasch, T., Werner, M., Fornasier, M., Stevenson, R.: Adaptive frame methods for elliptic operator equations: the steepest descent approach. IMA J. Numer. Anal. 27(4), 717–740 (2007)

    MathSciNet  MATH  Article  Google Scholar 

  12. 12.

    Dahmen, W.: Wavelet and multiscale methods for operator equations. Acta Numer. 6, 55–228 (1997)

    MathSciNet  MATH  Article  Google Scholar 

  13. 13.

    Dahmen, W., Huang, C., Kutyniok, G., Lim, W.-Q, Schwab, C., Welper, G.: Efficient resolution of anisotropic structures. In: Extraction of quantifiable information from complex systems, pp. 25–51. Springer (2014)

  14. 14.

    Dahmen, W., Huang, C., Schwab, C., Welper, G.: Adaptive petrov–galerkin methods for first order transport equations. SIAM J. Numer Anal. 50(5), 2420–2445 (2012)

    MathSciNet  MATH  Article  Google Scholar 

  15. 15.

    Dahmen, W., Kunoth, A., Urban, K.: A wavelet Galerkin method for the Stokes equations. Computing 56(3), 259–301 (1996)

    MathSciNet  MATH  Article  Google Scholar 

  16. 16.

    Dahmen, W., Kutyniok, G., Lim, W., Schwab, C., Welper, G.: Adaptive anisotropic Petrov-Galerkin methods for first order transport equations. J. Comput. Appl. Math. 340, 191–220 (2018)

    MathSciNet  MATH  Article  Google Scholar 

  17. 17.

    Dahmen, W., Schneider, R.: Wavelets with complementary boundary conditions—function spaces on the cube. Results Math. 34(3-4), 255–293 (1998)

    MathSciNet  MATH  Article  Google Scholar 

  18. 18.

    Daubechies, I.: Ten lectures on wavelets, volume 61 of CBMS-NSF Regional Conference Series in Applied Mathematics. Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA (1992)

  19. 19.

    Dauge, M., Stevenson, R.: Sparse tensor product wavelet approximation of singular functions. SIAM J. Math Anal. 42(5), 2203–2228 (2010)

    MathSciNet  MATH  Article  Google Scholar 

  20. 20.

    Demanet, L., Ying, L.: Wave atoms and time upscaling of wave equations. Numer. Math. 113(1), 1–71 (2009)

    MathSciNet  MATH  Article  Google Scholar 

  21. 21.

    Dijkema, T.J., Stevenson, R.: A sparse Laplacian in tensor product wavelet coordinates. Numer. Math. 115(3), 433–449 (2010)

    MathSciNet  MATH  Article  Google Scholar 

  22. 22.

    Donoho, D.: Sparse components of images and optimal atomic decompositions. Constr. Approx. 17(3), 353–382 (2001)

    MathSciNet  MATH  Article  Google Scholar 

  23. 23.

    Donoho, D.L., Kutyniok, G.: Microlocal analysis of the geometric separation problem. Comm Pure Appl. Math. 66, 1–47 (2013)

    MathSciNet  MATH  Article  Google Scholar 

  24. 24.

    Donovan, G.C., Geronimo, J.S., Hardin, D.P.: Intertwining multiresolution analyses and the construction of piecewise-polynomial wavelets. SIAM J. Math. Anal. 27(6), 1791–1815 (1996)

    MathSciNet  MATH  Article  Google Scholar 

  25. 25.

    Donovan, G.C., Geronimo, J.S., Hardin, D.P.: Orthogonal polynomials and the construction of piecewise polynomial smooth wavelets. SIAM J. Math. Anal. 30(5), 1029–1056 (1999)

    MathSciNet  MATH  Article  Google Scholar 

  26. 26.

    Easley, G., Labate, D., Negi, P.: 3D data denoising using combined sparse dictionaries. Math. Model. Nat Phenom. 8(1), 60–74 (2013)

    MathSciNet  MATH  Article  Google Scholar 

  27. 27.

    Elad, M., Starck, J.-L., Querre, P., Donoho, D.L.: Simultaneous cartoon and texture image inpainting using morphological component analysis (MCA). Appl. Comput. Harmon. Anal. 19, 340–358 (2005)

    MathSciNet  MATH  Article  Google Scholar 

  28. 28.

    Etter, S., Grohs, P., Obermeier, A.: FFRT - a fast finite ridgelet transform for radiative transport. Multiscale Model Simul. 13(1), 1–42 (2014)

    MathSciNet  Article  Google Scholar 

  29. 29.

    Genzel, M., Kutyniok, G.: Asymptotic analysis of inpainting via universal shearlet systems. SIAM J. Imaging Sci. 7, 2301–2339 (2014)

    MathSciNet  MATH  Article  Google Scholar 

  30. 30.

    Grohs, P., Obermeier, A.: Optimal adaptive ridgelet schemes for linear advection equations. Appl. Comput Harmon. Anal. 41(3), 768–814 (2016)

    MathSciNet  MATH  Article  Google Scholar 

  31. 31.

    Guo, K., Kutyniok, G., Labate, D. : Sparse multidimensional representations using anisotropic dilation and shear operators. In: Wavelets and splines: Athens 2005, Mod. Methods Math, pp 189–201. Nashboro Press, Brentwood (2006) ,

  32. 32.

    Guo, K., Labate, D.: Optimally sparse multidimensional representation using shearlets. SIAM J. Math. Anal. 39(1), 298–318 (2007)

    MathSciNet  MATH  Article  Google Scholar 

  33. 33.

    Harbrecht, H., Stevenson, R.: Wavelets with patchwise cancellation properties. Comp. Math. 75(256), 1871–1889 (2006)

    MathSciNet  MATH  Article  Google Scholar 

  34. 34.

    Kittipoom, P., Kutyniok, G.: Construction of compactly supported shearlet frames. Constr. Approx. 35(1), 21–72 (2012)

    MathSciNet  MATH  Article  Google Scholar 

  35. 35.

    Kutyniok, G., Lim, W.-Q.: Compactly supported shearlets are optimally sparse. J. Approx. Theory 163(11), 1564–1589 (2011)

    MathSciNet  MATH  Article  Google Scholar 

  36. 36.

    Kutyniok, G., Lim, W.-Q., Reisenhofer, R.: ShearLab 3D faithful digital shearlet transforms based on compactly supported shearlets. ACM Trans. Math. Softw. 42(1), 1–42 (2016)

    MathSciNet  MATH  Article  Google Scholar 

  37. 37.

    Labate, D., Lim, W.-Q., Kutyniok, G., Weiss, G.: Sparse multidimensional representation using shearlets. In: Proceedings SPIE (Wavelets and Sparsity XI), vol. 5914 (2005)

  38. 38.

    Nitsche, P.-A.: Best N term approximation spaces for tensor product wavelet bases. Constr. Approx. 24(1), 49–70 (2006)

    MathSciNet  MATH  Article  Google Scholar 

  39. 39.

    Petersen, P.: Shearlet approximation of functions with discontinuous derivatives. J. Approx. Theory 207, 127–138 (2016)

    MathSciNet  MATH  Article  Google Scholar 

  40. 40.

    Petersen, P.: Shearlets on Bounded Domains and Analysis of Singularities Using Compactly Supported Shearlets. Dissertation Technische Universiät Berlin (2016)

  41. 41.

    Petersen, P., Raslan, M.: Approximation properties of hybrid shearlet-wavelet frames for Sobolev spaces. Adv. Comp. Math., 2019, in press

  42. 42.

    Primbs, M.: Stabile biorthogonale Spline-Waveletbasen auf dem Intervall. Duisburg, Essen Univ., Dissertation (2006)

  43. 43.

    Stein, E.M.: Singular integrals and differentiability properties of functions. Princeton Mathematical Series, vol. 30. Princeton University Press, Princeton (1970)

    Google Scholar 

  44. 44.

    Stevenson, R.: Adaptive solution of operator equations using wavelet frames. SIAM J. Numer. Anal 41(3), 1074–1100 (2003)

    MathSciNet  MATH  Article  Google Scholar 

  45. 45.

    Voigtlaender, F.: Structured, compactly supported Banach frame decompositions of decomposition spaces. arXiv:1602.08772 (2016)

Download references

Acknowledgments

P. Petersen would like to thank Kristof Schröder and Massimo Fornasier for various discussions on related topics. Furthermore, P. Petersen expresses his gratitude to Massimo Fornasier and the Technische Universität München for the hospitality during P. Petersen’s research visit. Parts of this work was also done when J. Ma visited the Department of Applied Mathematics and Theoretical Physics of the University of Cambridge, and he is grateful for its hospitality. Parts of this work was accomplished while G. Kutyniok was visiting ETH Zürich. She is grateful to the Institute for Mathematical Research (FIM) and the Seminar for Applied Mathematics (SAM) for their hospitality and support during this visit.

Funding

J. Ma and P. Petersen acknowledge support from the DFG Collaborative Research Center TRR 109 “Discretization in Geometry and Dynamics”. G. Kutyniok acknowledges support by the Einstein Foundation Berlin, by the Einstein Center for Mathematics Berlin (ECMath), by Deutsche Forschungsgemeinschaft (DFG) SPP 1798, by the DFG Collaborative Research Center TRR 109 “Discretization in Geometry and Dynamics”, by the DFG Research Center Matheon “Mathematics for key technologies” in Berlin, and by the European Commission-Project DEDALE (contract no.665044) within the H2020 Framework Program.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Mones Raslan.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Communicated by: Ivan Oseledets

Appendices

Appendix A: \(H^{s}(\mathbb {R}^{2})\) Frame property of a reweighted shearlet system

This section shall be concerned with the proof of Theorem 3.3. The proof is split into Lemma A.2 for the upper frame bound and Lemma A.3 for the lower frame bound.

We use the following lemma to estimate weighted \(L^{2}(\mathbb {R}^{2})\) - frame coefficients by the \(H^{s}(\mathbb {R}^{2})\) norm of a function f.

Lemma A.1 (40, Lemma 2.4.9.)

Let\(s \in \mathbb {N}\), \(\tilde {C}>0,\)βsh > 3 and αsh > βsh + s. Further, let\(\phi , \psi , \widetilde {\psi } \in L^{2}(\mathbb {R}^{2})\)such that for almost all\(\xi \in \mathbb {R}^{2}\)there holds

$$ \begin{array}{@{}rcl@{}} |\hat \phi(\xi)| &\leq& \tilde{C} \min\{1, |{\xi}_{1} |^{-{\beta}_{\text{sh}}}\}\min\{1, |{\xi}_{2} |^{-{\beta}_{\text{sh}}}\}\\ |\hat \psi(\xi)| &\leq& \tilde{C} \min\{1, |{\xi}_{1} |^{{\alpha}_{\text{sh}}}\}\min\{1, |{\xi}_{1} |^{-{\beta}_{\text{sh}}}\}\min\{1, |{\xi}_{2} |^{-{\beta}_{\text{sh}}}\} \\ |\hat {\tilde{\psi}}(\xi)| &\leq& \tilde{C} \min\{1, | {\xi}_{2} |^{{\alpha}_{\text{sh}}}\}\min\{1, |{\xi}_{1} |^{-{\beta}_{\text{sh}}}\}\min\{1, |{\xi}_{2} |^{-{\beta}_{\text{sh}}}\}. \end{array} $$

Then there exists a constant C > 0 depending only on s, αshand βshsuch that\(\mathcal {SH}(\phi , \psi , \tilde {\psi }, c)\)satisfies

$$ \left\|\left( 2^{js}\langle f, {\psi}_{j,k,m,\iota}\rangle_{L^{2}(\mathbb{R}^{2})}\right)_{(j,k,m,\iota)\in {\Lambda}} \right\|_{\ell^{2}({\Lambda})}\leq \frac{C \tilde{C} }{\sqrt{|\det(M_{c})|}} \|f\|_{H^{s}(\mathbb{R}^{2})}, \quad \text{ for all } f \in H^{s}(\mathbb{R}^{2}). $$

We continue to give the upper frame bound for the weighted shearlet system in \(H^{s}(\mathbb {R}^{2})\).

Lemma A.2

For\(s\in \mathbb {N}\)let\(\phi ^{1}, {\psi }^{1} \in L^{2}(\mathbb {R})\)be such that for some\(\tilde {C} \geq 0\), αshs > βsh > 3, and all 0 ≤ r1, r2 ≤ 2s,

$$ \begin{array}{@{}rcl@{}} \left|\widehat{D^{r_{1}}\phi^{1}}(\xi)\right| &\leq& \sqrt{\tilde{C}} \min \{1,|\xi|^{-{\beta}_{\text{sh}}} \}, \text{ for all } \xi \in \mathbb{R}, \\ \left|\widehat{D^{r_{2}}{\psi}^{1}}(\xi)\right| &\leq& \sqrt{\tilde{C}} \min \{1,|\xi|^{{\alpha}_{\text{sh}}} \} \min \{1,|\xi|^{-{\beta}_{\text{sh}}} \}, \text{ for all } \xi \in \mathbb{R}. \end{array} $$
(12)

We further denote ϕ = ϕ1ϕ1 and ψ = ψ1ϕ1and\(\tilde {\psi }(x_{1}, x_{2}) := \psi (x_{2}, x_{1})\)for all\(x\in \mathbb {R}^{2}\). Then, there exists some B > 0 depending only on αsh, βshand ssuch for\(({\psi }_{j,k,m,\iota })_{(j,k,m,\iota )\in {\Lambda }} = \mathcal {SH}(\phi , \psi , \tilde {\psi }, (c_{1},c_{2}))\)we have the estimate

$$ \begin{array}{@{}rcl@{}} \sum\limits_{(j,k,m,\iota)\in {\Lambda}} |\langle f, 2^{-js} {\psi}_{j,k,m,\iota} \rangle_{H^{s}(\mathbb{R}^{2})} |^{2} \leq B \frac{\tilde{C}}{|\det(M_{c})|} \|f\|_{H^{s}(\mathbb{R}^{2})}^{2} \end{array} $$
(13)

for all\(f \in H^{s}(\mathbb {R}^{2})\).

Proof

We define for \(\mathbf {r}= (r_{1},r_{2})\): \(\phi ^{(\mathbf {r})} = D^{r_{1}}\phi ^{1} \otimes D^{r_{2}}\phi ^{1}\), \({\psi }^{(\mathbf {r})} = D^{r_{1}}{\psi }^{1}\otimes D^{r_{2}}\phi ^{1}\), and \(\tilde {\psi }^{(\mathbf {r})}(x_{1},x_{2})={\psi }^{(\mathbf {r})}(x_{2},x_{1})\). By Lemma A.1 there exists a constant C, dependent only on \({\alpha }_{\text {sh}}, {\beta }_{\text {sh}},\) such that for all \(0\leq r_{1},r_{2} \leq 2s\) and for \(({\psi }_{j,k,m,\iota }^{(\mathbf {r})})_{(j,k,m,\iota )\in {\Lambda }} = \mathcal {SH}(\phi ^{(\mathbf {r})}, {\psi }^{(\mathbf {r})}, \tilde {\psi }^{(\mathbf {r})}, c)\) we have that

$$ \begin{array}{@{}rcl@{}} \sum\limits_{(j,k,m,\iota)\in {\Lambda}} 2^{2js}|\langle f, {\psi}_{j,k,m,\iota}^{(\mathbf{r})}\rangle_{L^{2}(\mathbb{R}^{2})} |^{2} \leq C \frac{\tilde{C}}{|\det(M_{c})|} \|f\|_{H^{s}(\mathbb{R}^{2})}^{2}, \text{ for all } f\in H^{s}(\mathbb{R}^{2}). \end{array} $$
(14)

From the assumptions (12) we have that \(\phi , \psi , \tilde {\psi } \in H^{2s}(\mathbb {R}^{2})\) and thus we can estimate the left-hand side of (13).

$$ \begin{array}{@{}rcl@{}} &&\sum\limits_{(j,k,m,\iota) \in {\Lambda}} | 2^{-js}\left \langle f, {\psi}_{j,k,m,\iota}\right \rangle_{H^{s}(\mathbb{R}^{2})} |^{2} \\ &=& \sum\limits_{(j,k,m,\iota) \in {\Lambda}} | \sum\limits_{|\mathbf{a}|\leq s} 2^{-js} \left \langle D^{\mathbf{a}}f, D^{\mathbf{a}}{\psi}_{j,k,m,\iota}\right \rangle_{L^{2}(\mathbb{R}^{2})} |^{2}\\ &\lesssim& \sum\limits_{(j,k,m,\iota) \in {\Lambda}} \sum\limits_{|\mathbf{a}|\leq s}|2^{-js} \left \langle f, D^{2\mathbf{a}}{\psi}_{j,k,m,\iota}\right \rangle_{L^{2}(\mathbb{R}^{2})} |^{2} = \mathrm{I}. \end{array} $$
(4)

By using the product rule we calculate

$$ \begin{array}{@{}rcl@{}} D^{2\mathbf{a}}{\psi}_{j,k,m,\iota} = \sum\limits_{0\leq r_{1},r_{2} \leq 2s} c_{\mathbf{r}}^{(\mathbf{a})} {\psi}^{(\mathbf{r})}_{j,k,m,\iota}, \end{array} $$
(16)

with coefficients \(c_{\mathbf {r}}^{\mathbf {a}}\) bounded in absolute value by 22js. After applying (16) to (15) we estimate

$$ \begin{array}{@{}rcl@{}} \mathrm{I} &\lesssim& \sum\limits_{(j,k,m,\iota) \in {\Lambda}} \sum\limits_{|\mathbf{a}|\leq s} \sum\limits_{0\leq r_{1}, r_{2} \leq 2s} 2^{2js} | \langle f, {\psi}_{j,k,m,\iota}^{(\mathbf{r})}\rangle_{L^{2}(\mathbb{R}^{2})}|^{2}\\ &\lesssim& \sup\limits_{0\leq r_{1},r_{2} \leq 2s}\sum\limits_{(j,k,m,\iota) \in {\Lambda}} 2^{2js} | \langle f, {\psi}_{j,k,m,\iota}^{(\mathbf{r})} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2} =\text{II}. \end{array} $$

Invoking (14), there exists some B > 0 such that

$$ \begin{array}{@{}rcl@{}} \text{II} \leq B \frac{\tilde{C}}{|\det(M_{c})|} \|f\|_{H^{s}(\mathbb{R}^{2})}^{2}, \text{ for all } f\in H^{s}(\mathbb{R}^{2}). \end{array} $$

This yields the result. □

As a next step we provide the lower frame bound.

Lemma A.3

For\(s\in \mathbb {N}\), αshs > βsh > 4 and some\(\tilde {C}>0\)let\(\phi ^{1}, {\psi }^{1} \in L^{2}(\mathbb {R})\)be such that for all 0 ≤ r1, r2 ≤ 2s,

$$ \begin{array}{@{}rcl@{}} \left|\widehat{D^{r_{1}}\phi^{1}}(\xi)\right| &\leq& \sqrt{\tilde{C}}\min \{1,|\xi|^{-{\beta}_{\text{sh}}} \}, \text{ for all } \xi \in \mathbb{R}, \\ \left|\widehat{D^{r_{2}}{\psi}^{1}}(\xi)\right| &\leq& \sqrt{\tilde{C}} \min \{1,|\xi|^{{\alpha}_{\text{sh}}} \} \min \{1,|\xi|^{-{\beta}_{\text{sh}}} \}, \text{ for all } \xi \in \mathbb{R}. \end{array} $$

Further, let\(\phi , \psi , \widetilde {\psi }, \theta , \widetilde {\theta }, \mu \)be as in (1). Assume that there exists\(\bar c>0\)such that for all\(c_{1},c_{2}\leq \bar {c}\)we have that\(\mathcal {SH}(\mu ,\theta ,\tilde {\theta },(c_{1},c_{2}))\)forms a frame for\(L^{2}(\mathbb R^{2})\)with lower frame bound, which can be bounded from below independently ofc1, c2. Then there exists\(\tilde {c}>0\)such that for all\(c_{1}=c_{2} \leq \tilde {c}\)and c = (c1, c2) the system\(({\psi }_{j,k,m,\iota })_{(j,k,m,\iota )\in {\Lambda }} = \mathcal {SH}(\mu ,\theta ,\tilde {\theta },c)\), obeys

$$ \|f\|_{H^{s}(\mathbb{R}^{2})}^{2}\lesssim \sum\limits_{(j,k,m,\iota)\in {\Lambda}} |\langle f, 2^{-js} {\psi}_{j,k,m,\iota} \rangle_{H^{s}(\mathbb{R}^{2})} |^{2}, $$

for all\(f\in H^{s}(\mathbb {R}^{2})\).

Proof

Let \(\bar {c} \geq c>0\). We define

$$ \hat{\mu}^{c}:=\mathcal{X}_{\left[-\frac{1}{2c_{1}},\frac{1}{2c_{1}}\right]^{2}}\hat{\mu}, \qquad \hat{\theta}^{c}:=\mathcal{X}_{\left[-\frac{1}{2c_{1}},\frac{1}{2c_{1}}\right]^{2}}\hat{\theta} \qquad, \hat{\widetilde{\theta}}^{c}:=\mathcal{X}_{\left[-\frac{1}{2c_{1}},\frac{1}{2c_{1}}\right]^{2}}\hat{\widetilde{\theta}}. $$

By using the fact, that \(\mathcal {SH}(\mu ,\theta ,\tilde {\theta },c)\) is a frame for \(L^{2}(\mathbb {R}^{2}),\) we deduce

$$ \begin{array}{@{}rcl@{}} \|f\|_{H^{s}(\mathbb{R}^{2})}^{2}\ &\sim& \| (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}\|_{L^{2}(\mathbb{R}^{2})}^{2}\\ & \lesssim& \sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}, \widehat{\theta}_{j,k,m,\iota} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2} \\ &\leq& \sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}, \widehat{\theta}^{c}_{j,k,m} \rangle_{L^{2}(\mathbb{R}^{2})}|^{2} \\ &&+ \sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}, \widehat{\theta}_{j,k,m}-\widehat{\theta}^{c}_{j,k,m} \rangle_{L^{2}(\mathbb{R}^{2})}|^{2} = \mathrm{I}_{1} + \mathrm{I}_{2}. \end{array} $$
(17)

Since βsh > 4, there exists 𝜖 > 0 such that βsh − 1 − 𝜖 > 3 and for all \(\xi \in \mathbb {R}^{2}\)

$$ \begin{array}{@{}rcl@{}} \min \{1, |\xi|^{{\alpha}_{\text{sh}}}\} \min \{ 1, |\xi|^{-{\beta}_{\text{sh}}}\} \leq \min \{1, |{\xi}_{1}|^{{\alpha}_{\text{sh}}}\} \min \{ 1, |{\xi}_{1}|^{-{\beta}_{\text{sh}}+1+\epsilon}\} (\max \{ 1, |{\xi}_{1}|\})^{-1-\epsilon}. \end{array} $$

Thus, we have that \((\theta _{j,k,m,\iota }- \theta _{j,k,m,\iota }^{c})_{(j,k,m,\iota ) \in {\Lambda }}\) satisfies the assumptions of Lemma A.2 with \(\sqrt {\tilde {C}} = 2c_{1}^{1+\epsilon }\). Hence, I2 can be estimated by \(c_{1}^{2+2\epsilon }/|\det {(M_{c})}| \|f\|_{H^{s}(\mathbb {R}^{2})}^{2}\) and since \(\det {(M_{c}}) = {c_{1}^{2}}\) this term is negligible for c1 small enough. Therefore we obtain

$$ \begin{array}{@{}rcl@{}} \|f\|_{H^{s}(\mathbb{R}^{2})}^{2} & \lesssim& \sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}, \widehat{\theta}^{c}_{j,k,m} \rangle_{L^{2}(\mathbb{R}^{2})}|^{2}\\ &=& \sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}, \widehat{\theta}^{c}_{j,k,m} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2}\\ &&+ \sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f},\widehat{\widetilde{\theta}}^{c}_{j,k,m} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2} \\ &&+ \sum\limits_{m\in\mathbb{Z}^{2}} |\langle (1+|\cdot|^{2})^{\frac{s}{2}} \hat{f}, \mathcal{F}(\mu^{c}(\cdot-c_{1}m)) \rangle_{L^{2}(\mathbb{R}^{2})} |^{2} \end{array} $$
$$ \begin{array}{@{}rcl@{}} &=& \sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} 2^{\frac{3j}{2}}\left\vert {\int}_{\mathbb{R}^{2}} \hat{f}(\xi)(1+|\xi|^{2})^{\frac{s}{2}} \overline{\mathcal{F}(\theta^{c}(S_{k}A_{j}\cdot-M_{c}m))(\xi)}d\xi \right\vert^{2} \\ &&+\sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} 2^{\frac{3j}{2}}\left\vert {\int}_{\mathbb{R}^{2}} \hat{f}(\xi)(1+|\xi|^{2})^{\frac{s}{2}} \overline{\mathcal{F}(\widetilde{\theta}^{c}({S_{k}^{T}}\tilde{A}_{j}\cdot-M_{\tilde{c}}m))(\xi)}d\xi \right\vert^{2} \\ &&+\sum\limits_{m\in\mathbb{Z}^{2}} \left\vert{\int}_{\mathbb{R}^{2}}\hat{f}(\xi)(1+|\xi|^{2})^{\frac{s}{2}}\overline{\mathcal{F}(\mu^{c}(\cdot-c_{1}m))(\xi)} d\xi\right\vert^{2} \\ &=& \sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} 2^{-\frac{3j}{2}}\left\vert {\int}_{\mathbb{R}^{2}} \hat{f}(\xi)(1+|\xi|^{2})^{\frac{s}{2}} \overline{\hat{\theta}^{c}(S_{k}^{-T}A_{j}^{-1}\xi)}e^{2\pi i\langle M_{c}S_{k}^{-T}A_{j}^{-1}\xi, m \rangle}d\xi \right\vert^{2} \\ &&+\sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} 2^{-\frac{3j}{2}}\left\vert {\int}_{\mathbb{R}^{2}} \hat{f}(\xi)(1+|\xi|^{2})^{\frac{s}{2}} \overline{\hat{\widetilde{\theta}}^{c}(S_{k}^{-1}\tilde{A}_{j}^{-1}\xi)} e^{2\pi i\langle M_{\tilde{c}}S_{k}^{-1}\tilde{A}_{j}^{-1}\xi,m\rangle}d\xi \right\vert^{2} \\ &&+\sum\limits_{m\in\mathbb{Z}^{2}} \left\vert{\int}_{\mathbb{R}^{2}}\hat{f}(\xi)(1+|\xi|^{2})^{\frac{s}{2}}\overline{\hat{\mu}^{c}(\xi)}e^{2\pi i\langle c_{1}\xi,m \rangle} d\xi\right\vert^{2} = \text{II}. \end{array} $$

Now we substitute

$$ \begin{array}{@{}rcl@{}} \xi\leadsto A_{j}{S_{k}^{T}}M_{c}^{-1}\xi, \qquad \xi\leadsto \tilde{A}_{j}S_{k}M_{\tilde{c}}^{-1}\xi, \qquad \xi\leadsto \frac{\xi}{c_{1}}. \end{array} $$
(18)

Furthermore, we use that μc, 𝜃c, and \(\tilde {\theta }^{c}\) are all supported in \([-\frac {1}{2c_{1}},\frac {1}{2c_{1}}]^{2}\).

$$ \begin{array}{@{}rcl@{}} \text{II}&=& \sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} \frac{2^{\frac{3j}{2}}}{|\det(M_{c})|^{2}}\cdot \left\vert {\int}_{[-\frac{1}{2},\frac{1}{2}]^{2}} \hat{f}(A_{j}{S_{k}^{T}}M_{c}^{-1}\xi)(1+|A_{j}{S_{k}^{T}}M_{c}^{-1}\xi|^{2})^{\frac{s}{2}}\overline{\hat{\theta}^{c}(M_{c}^{-1}\xi)}e^{2\pi i\langle \xi,m \rangle} d\xi\right\vert^{2} \\ &&+ \sum\limits_{(j,k,m)\in{\Lambda}^{\prime}} \frac{2^{\frac{3j}{2}}}{|\det(M_{c})|^{2}}\cdot \left\vert {\int}_{[-\frac{1}{2},\frac{1}{2}]^{2}} \hat{f}(\tilde{A}_{j}S_{k}M_{\tilde{c}}^{-1}\xi)(1+|\tilde{A}_{j}S_{k} M_{\tilde{c}}^{-1}\xi|^{2})^{\frac{s}{2}}\overline{\hat{\widetilde{\theta}}^{c}(M_{\tilde{c}}^{-1}\xi)}e^{2\pi i\langle \xi,m \rangle} d\xi\right\vert^{2} \\ &&+ \sum\limits_{m\in\mathbb{Z}^{2}} \frac{1}{{c_{1}^{2}}}\left\vert {\int}_{[-\frac{1}{2},\frac{1}{2}]^{2}} \hat{f}\left( \frac{\xi}{c_{1}} \right) \left( 1+\left\vert \frac{\xi}{c_{1}} \right\vert^{2} \right)^{\frac{s}{2}} \overline{\hat{\mu}^{c}\left( \frac{\xi}{c_{1}}\right)}e^{2\pi i\langle\xi,m\rangle}d\xi\right\vert^{2}. \end{array} $$

By using the Parseval identity we obtain

$$ \begin{array}{@{}rcl@{}} \text{II}&=& \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} \frac{2^{\frac{3j}{2}}}{|\det(M_{c})|^{2}}\cdot \left\| \hat{f}(A_{j}{S_{k}^{T}}M_{c}^{-1}\cdot)(1+|A_{j}{S_{k}^{T}}M_{c}^{-1}\cdot|^{2})^{\frac{s}{2}}\hat{\theta}^{c}(M_{c}^{-1}\cdot)\right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} \frac{2^{\frac{3j}{2}}}{|\det(M_{c})|^{2}}\cdot \left\| \hat{f}(\tilde{A}_{j}S_{k}M_{\tilde{c}}^{-1}\cdot)(1+|\tilde{A}_{j}S_{k} M_{\tilde{c}}^{-1}\cdot|^{2})^{\frac{s}{2}}\hat{\widetilde{\theta}}^{c}(M_{\tilde{c}}^{-1}\cdot)\right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \frac{1}{{c_{1}^{2}}}\left\| \hat{f}\left( \frac{\cdot}{c_{1}} \right) \left( 1+\left\vert \frac{\cdot}{c_{1}} \right\vert^{2} \right)^{\frac{s}{2}} \hat{\mu}^{c}\left( \frac{\cdot}{c_{1}}\right)\right\|_{L^{2}(\mathbb{R}^{2})}^{2}. \end{array} $$

Now we substitute

$$ \begin{array}{@{}rcl@{}} \xi\leadsto M_{c}\xi, \qquad \xi\leadsto M_{\tilde{c}}\xi, \qquad \xi\leadsto c_{1}\xi \end{array} $$
(19)

to arrive at

$$ \begin{array}{@{}rcl@{}} \text{II}&=& \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}} \left\| (1+|A_{j}{S_{k}^{T}}\cdot|^{2})^{\frac{s}{2}}\hat{f}(A_{j}{S_{k}^{T}}\cdot) \hat{\theta}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}}2^{\frac{3j}{2}} \left\| (1+|\tilde{A}_{j}S_{k}\cdot|^{2})^{\frac{s}{2}} \hat{f}(\tilde{A}_{j}S_{k}\cdot) \hat{\widetilde{\theta}}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \left\| (1+|\cdot|^{2})^{\frac{s}{2}}\cdot\hat{f}\cdot \hat{\phi} \right\|_{L^{2}(\mathbb{R}^{2})}^{2}. \end{array} $$

Now set

$$ \hat{\phi}^{c}:=\mathcal{X}_{[-\frac{1}{2c_{1}},\frac{1}{2c_{1}}]^{2}}\hat{\phi}, \qquad \hat{\psi}^{c}:=\mathcal{X}_{[-\frac{1}{2c_{1}},\frac{1}{2c_{1}}]^{2}}\hat{\psi} \qquad, \hat{\widetilde{\psi}}^{c}:=\mathcal{X}_{[-\frac{1}{2c_{1}},\frac{1}{2c_{1}}]^{2}}\hat{\widetilde{\psi}}. $$

Using the definition of \(\theta ^{c},\widetilde {\theta }^{c},\mu ^{c}\) then yields

$$ \begin{array}{@{}rcl@{}} \text{II}&=&\sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}} \left\| (1+|A_{j}{S_{k}^{T}}\cdot|^{2})^{\frac{s}{2}}\hat{f}(A_{j}{S_{k}^{T}}\cdot) |(\cdot)_{1}|^{s} \hat{\psi}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}}2^{\frac{3j}{2}} \left\| (1+|\tilde{A}_{j}S_{k}\cdot|^{2})^{\frac{s}{2}} \hat{f}(\tilde{A}_{j}S_{k}\cdot) |(\cdot)_{2}|^{s} \hat{\widetilde{\psi}}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \left\| (1+|\cdot|^{2})^{\frac{s}{2}}\cdot\hat{f}\cdot \hat{\phi} \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &=& \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}} \left\| (1+|A_{j}{S_{k}^{T}}\cdot|^{2})^{\frac{s}{2}}\hat{f}(A_{j}{S_{k}^{T}}\cdot) 2^{-js} |(A_{j}{S_{k}^{T}}\cdot)_{1}|^{s} \hat{\psi}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}}2^{\frac{3j}{2}} \left\| (1+|\tilde{A}_{j}S_{k}\cdot|^{2})^{\frac{s}{2}} \hat{f}(\tilde{A}_{j}S_{k}\cdot) 2^{-js} |(\tilde{A}_{j}S_{k}\cdot)_{2}|^{s} \hat{\psi}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \left\| (1+|\cdot|^{2})^{\frac{s}{2}}\cdot\hat{f}\cdot \hat{\phi}^{c} \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &\leq& \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}}\left\| (1+|A_{j}{S_{k}^{T}}\cdot|^{2})^{s}\hat{f}(A_{j}{S_{k}^{T}}\cdot) 2^{-js} \hat{\psi}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}}\left\| (1+|\tilde{A}_{j}S_{k}\cdot|^{2})^{s} \hat{f}(\tilde{A}_{j}S_{k}\cdot) 2^{-js} \hat{\widetilde{\psi}}^{c}(\cdot) \right\|_{L^{2}(\mathbb{R}^{2})}^{2} \\ &&+ \left\| (1+|\cdot|^{2})^{s}\cdot\hat{f}\cdot \hat{\phi}^{c} \right\|_{L^{2}(\mathbb{R}^{2})}^{2} = \text{III}. \end{array} $$

We observe that

$$ (1+|\xi|^{2})^{s} \lesssim \sum\limits_{|\mathbf{a}|\leq s} (2\pi)^{2|\mathbf{a}|} \xi^{2\mathbf{a}} \text{ for all } \xi \in \mathbb{R}^{2}. $$

Thus we can estimate

$$ \begin{array}{@{}rcl@{}} \text{III} \lesssim &\sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}}\cdot\left\|\sum\limits_{|\mathbf{a}|\leq s} (2\pi)^{2|\mathbf{a}|} (A_{j}{S_{k}^{T}}\cdot)^{2\mathbf{a}} \hat{f}(A_{j}{S_{k}^{T}}\cdot)2^{-js} \hat{\psi}^{c}(\cdot)\right\|_{L^{2}(\mathbb{R}^{2})}^{2}\\ &\quad + \sum\limits_{(j,k)\in{\Lambda}^{\prime\prime}} 2^{\frac{3j}{2}}\cdot\left\|\sum\limits_{|\mathbf{a}|\leq s} (2\pi)^{2|\mathbf{a}|} (\tilde{A}_{j}S_{k}\cdot)^{2\mathbf{a}}\hat{f}(\tilde{A}_{j}S_{k}\cdot) 2^{-js} \hat{\widetilde{\psi}}^{c}(\cdot)\right\|_{L^{2}(\mathbb{R}^{2})}^{2}\\ &\qquad + \left\|\sum\limits_{|\mathbf{a}|\leq s} (2\pi)^{2|\mathbf{a}|} (\cdot)^{2\mathbf{a}} \hat{f}\hat{\phi}^{c} \right\|_{L^{2}(\mathbb{R}^{2})}^{2} = \text{IV}. \end{array} $$

Invoking Parseval’s identity again and reversing the transformations (19) as well as (18) from earlier shows that

$$ \begin{array}{@{}rcl@{}} \text{IV} &= &\sum\limits_{(j,k,m,\iota)\in {\Lambda}} |\langle \hat{f}, \sum\limits_{|\mathbf{a}|\leq s} (2\pi)^{2|\mathbf{a}|} (\cdot)^{2\mathbf{a}} 2^{-js}\widehat{\psi}^{c}_{j,k,m,\iota} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2} \\ &= &\sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\sum\limits_{|\mathbf{a}|\leq s} \langle \hat{f}, (-1)^{|\mathbf{a}|} (2\pi i)^{2|\mathbf{a}|} (\cdot)^{2\mathbf{a}} 2^{-js}\widehat{\psi}^{c}_{j,k,m,\iota} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2}. \end{array} $$

By standard results on derivatives and the Fourier transform we have that

$$ (2\pi i \cdot)^{2\mathbf{a}} 2^{-js}\widehat{\psi}^{c}_{j,k,m,\iota} = \mathcal{F}(2^{-js} D^{2\mathbf{a}} {\psi}^{c}_{j,k,m,\iota}). $$

Thus we can invoke the Plancherel identity and partial integration to obtain

$$ \begin{array}{@{}rcl@{}} \text{IV}&= &\sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\sum\limits_{|\mathbf{a}|\leq s} \langle f, (-1)^{|\mathbf{a}|} 2^{-js}D^{2\mathbf{a}} {\psi}^{c}_{j,k,m,\iota} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2}\\ &= &\sum\limits_{(j,k,m,\iota)\in{\Lambda}} |\sum\limits_{|\mathbf{a}|\leq s} \langle D^{\mathbf{a}} f, 2^{-js}D^{\mathbf{a}} {\psi}^{c}_{j,k,m,\iota} \rangle_{L^{2}(\mathbb{R}^{2})} |^{2}\\ &= &\sum\limits_{(j,k,m,\iota)\in{\Lambda}} | \langle f, 2^{-js} {\psi}^{c}_{j,k,m,\iota} \rangle_{H^{s}(\mathbb{R}^{2})} |^{2}. \end{array} $$

We still need to transition back from \({\psi }^{c}_{j,k,m,\iota }\) to ψj, k, m, ι. We proceed by invoking a triangle inequality

$$ \begin{array}{@{}rcl@{}} \text{IV} &\lesssim & \ \sum\limits_{(j,k,m,\iota)\in{\Lambda}} | \langle f, 2^{-js} {\psi}_{j,k,m,\iota} \rangle_{H^{s}(\mathbb{R}^{2})} |^{2} \\ &&+ \sum\limits_{(j,k,m,\iota)\in{\Lambda}} | \langle f, 2^{-js} ({\psi}_{j,k,m,\iota} - {\psi}^{c}_{j,k,m,\iota}) \rangle_{H^{s}(\mathbb{R}^{2})} |^{2}\\ &= & \ \text{IV}_{1} + \text{IV}_{2}, \end{array} $$

Using a similar estimate as for I2 in (17) we can estimate by invoking Lemma A.2 that IV2 is negligible for c small enough. This yields the result. □

To conclude this subsection we would like to examine how the auxiliary functions \(\theta ,\widetilde {\theta }\) and μ can be chosen such that they fulfill the frame property required in the proof of Theorem 3.3. Therefore assume \(\psi , \widetilde {\psi },\) and ϕ fulfill the assumptions of Theorem 3.3 with γ + 4 > αsh > γ > 4 + s. Then it follows that

$$ \begin{array}{@{}rcl@{}} \theta(x) &=& \frac{1}{(-2\pi i)^{s}} D^{s}({\psi}^{1})(x_{1}) \phi^{1}(x_{2}),\\ \widetilde{\theta}(x) &=& \phi^{1}(x_{1}) \frac{1}{(-2\pi i)^{s}} D^{s}({\psi}^{1})(x_{2}),\\ \mu &=& \phi^{1}(x_{1}) \phi^{1}(x_{2}). \end{array} $$

Furthermore Ds(ψ1) and ϕ1 satisfy the assumptions of Theorem 3.3 and therefore there exists a sampling parameter \(\bar {c}>0\) such that for \(c_{1}=c_{2} \leq \bar c\) and c = (c1, c2) the system \(\mathcal {SH}(\phi , \psi , \widetilde {\psi }, c)\) constitutes a frame for \(L^{2}(\mathbb {R}^{2})\).

Appendix B: Localization of shearlet and wavelet frames

We now turn to the proof of Proposition 3.2. For this, we will require the following technical lemma.

Lemma B.1

[40] Let\(\psi \in L^{2}(\mathbb {R}^{2})\)be such that there exists C > 0 with

$$ \begin{array}{@{}rcl@{}} |\widehat{\psi}({\xi}_{1},{\xi}_{2})| \leq C \frac{\min\{1,|{\xi}_{1}|^{\alpha }\}}{\max\{1, |{\xi}_{1}|^{\beta} \} \max\{1, |{\xi}_{2}|^{\beta} \}}, \quad \text{for a.e. } ({\xi}_{1}, {\xi}_{2}) \in \mathbb{R}^{2}, \end{array} $$

where β/2 > α > 1. Then, for ι = − 1, 1,

$$ \begin{array}{@{}rcl@{}} \sum\limits_{|k| \leq 2^{j/2}}|({\psi}_{j,k,m,\iota})^{\wedge}({\xi}_{1},{\xi}_{2})| \leq 2^{-3/4j} C^{\prime} \frac{1}{\max\{1,|2^{-j}{\xi}_{1}|^{\beta/2}\}}\frac{1}{\max\{1,|2^{-j}{\xi}_{2}|^{\beta/2}\}}, \end{array} $$

for a.e. \(({\xi }_{1}, {\xi }_{2}) \in \mathbb {R}^{2}\)and a constant\(C^{\prime }\).

Proof

See [40, Lemma 3.2.1] for a proof. □

Now we are ready to prove Proposition 3.2.

Proof

of Proposition 3.2 First of all, by the definition of \({\Theta }_{\tau , t}^{c}\) one has that for all \((j_{\text {sh}},k,m_{\text {sh}},\iota ) \in {{\Lambda }_{0}^{c}}\) and \((j_{\mathrm {w}},m_{\mathrm {w}},\upsilon ) \in {\Theta }_{\tau , t}^{c}\) such that \(j_{\mathrm {w}} < 1/(2\tau ) j_{\text {sh}} + t\) we have that \(supp (2^{-j_{\mathrm {w}}s}\omega _{j_{\mathrm {w}},m_{\mathrm {w}},\upsilon })^{d} \cap supp {\psi }_{j_{\text {sh}},k,m_{\text {sh}},\iota } = \emptyset \). Hence, we can assume in the sequel that \(j_{\mathrm {w}} > 1/(2\tau ) j_{\text {sh}} + t\).

Furthermore, the total number of wavelet translates for a fixed level jw is of order \(2^{2j_{\mathrm {w}}}\). W.l.o.g., the following computations can be done for υ = 1. Using this observation and Plancherel’s identity, we obtain

$$ \begin{array}{@{}rcl@{}} &&\sum\limits_{(j_{\mathrm{w}},m_{\mathrm{w}},1)\in {\Theta}_{\tau, t }^{c}} \sum\limits_{(j_{\text{sh}},k,m_{\text{sh}},\iota) \in {{\Lambda}_{0}^{c}}} |\langle (2^{-j_{\mathrm{w}}s}\omega_{j_{\mathrm{w}},m_{\mathrm{w}},{1}})^{d}, 2^{-j_{\text{sh}} s}{\psi}_{j_{\text{sh}},k,m_{\text{sh}},\iota} \rangle_{H^{s}({\Omega})}|^{2} \\ &\!\lesssim\! & \ \sum\limits_{j_{\mathrm{w}} = 0}^{\infty}\sum\limits_{j_{\text{sh}} =0}^{(2\tau)(j_{\mathrm{w}}-t)} \sum\limits_{|k| \leq 2^{j_{\text{sh}}/2} } 2^{2j_{\mathrm{w}}}\max\limits_{m_{\text{sh}}, m_{\mathrm{w}}}(\sum\limits_{|\mathbf{a}| \leq s}|\langle (\cdot)^{\mathbf{a}} \widehat{(2^{-j_{\mathrm{w}} s}\omega_{j_{\mathrm{w}},m_{\mathrm{w}},{1}})^{d}}, (\cdot)^{\mathbf{a}} 2^{-j_{\text{sh}} s}\widehat{{\psi}_{j_{\text{sh}},k,m_{\text{sh}},\iota}} \rangle_{L^{2}(\mathbb{R}^{2})}|^{2})\\ &\!\lesssim\! & \ \sum\limits_{|\mathbf{a}| \leq s}\sum\limits_{j_{\mathrm{w}} = 0}^{\infty}\sum\limits_{j_{\text{sh}} =0}^{(2\tau)(j_{\mathrm{w}}-t)} \sum\limits_{|k| \leq 2^{j_{\text{sh}}/2} } 2^{2j_{\mathrm{w}}}\max\limits_{m_{\text{sh}}, m_{\mathrm{w}}}(|\langle (\cdot)^{\mathbf{a}} \widehat{(2^{-j_{\mathrm{w}} s}\omega_{j_{\mathrm{w}},m_{\mathrm{w}},{1}})^{d}}, (\cdot)^{\mathbf{a}} 2^{-j_{\text{sh}} s}\widehat{{\psi}_{j_{\text{sh}},k,m_{\text{sh}},\iota}} \rangle_{L^{2}(\mathbb{R}^{2})}|^{2}). \end{array} $$
(9)

Leveraging on the frequency decay of the corresponding shearlet atoms, applying Lemma B.1, and using (W1) yields for any |a|≤ s

$$ \begin{array}{@{}rcl@{}} &&\sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} \sum\limits_{{|k|}\leq 2^{j_{\text{sh}}/2}} 2^{2j_{\mathrm{w}}}\max\limits_{m_{\mathrm{w}},m_{\text{sh}}} |\langle (\cdot)^{\mathbf{a}} \widehat{(2^{-j_{\mathrm{w}} s}\omega_{j_{\mathrm{w}},m_{\mathrm{w}},{1}})^{d}}, (\cdot)^{\mathbf{a}} 2^{-j_{\text{sh}} s} \widehat{{\psi}_{j_{\text{sh}},k,m_{\text{sh}},\iota}} \rangle_{L^{2}(\mathbb{R}^{2})}|^{2}\\ & \lesssim& \sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} 2^{- 3/2j_{\text{sh}}} \left( {\int}_{\mathbb{R}^{2}} \frac{2^{-j_{\mathrm{w}} s}\xi^{\mathbf{a}} \min\{1, |2^{-j_{\mathrm{w}}} {\xi}_{1}|^{{\alpha}_{\mathrm{w}}}\}}{\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{1}|^{{\beta}_{\mathrm{w}}}\}\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{2}|^{{\beta}_{\mathrm{w}}}\}} \right. \\ && \quad \cdot \left. \frac{ 2^{-j_{\text{sh}} s}\xi^{\mathbf{a}}}{\max\{1, |2^{- j_{\text{sh}}}{\xi}_{1}|^{{\beta}_{\text{sh}}/2}\}\max\{1, |2^{-j_{\text{sh}}}{\xi}_{2}|^{{\beta}_{\text{sh}}/2}\}} d\xi \right)^{2} =: \mathrm{I}. \end{array} $$

By a simple computation we obtain that if βws

$$ \begin{array}{@{}rcl@{}} \frac{2^{-j_{\mathrm{w}} s}\xi^{\mathbf{a}} }{\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{1}|^{{\beta}_{\mathrm{w}}}\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{2}|^{{\beta}_{\mathrm{w}}-s}\} }\lesssim \frac{1}{\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{1}|^{{\beta}_{\mathrm{w}}-s}\} \max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{2}|^{{\beta}_{\mathrm{w}}-s}\} }. \end{array} $$

We plug this estimate into the estimate above and obtain with \({\beta }_{\mathrm {w}}^{\prime }={\beta }_{\mathrm {w}}-s\) and \({\beta }_{\text {sh}}^{\prime }={\beta }_{\text {sh}}/2-s\):

$$ \begin{array}{@{}rcl@{}} \mathrm{I} & \lesssim& \sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} 2^{- 3/2j_{\text{sh}}} \left( {\int}_{\mathbb{R}^{2}} \frac{ \min\{1, |2^{-j_{\mathrm{w}}} {\xi}_{1}|^{{\alpha}_{\mathrm{w}}}\}}{\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{1}|^{{\beta}_{\mathrm{w}}^{\prime}}\}\max\{1, |2^{-j_{\mathrm{w}}}{\xi}_{2}|^{{\beta}_{\mathrm{w}}^{\prime}}\}} \right. \\ & &\quad \cdot \left. \frac{1}{\max\{1, |2^{- j_{\text{sh}}}{\xi}_{1}|^{{\beta}_{\text{sh}}^{\prime}}\}\max\{1, |2^{-j_{\text{sh}}}{\xi}_{2}|^{{\beta}_{\text{sh}}^{\prime}}\}} d\xi \right)^{2} = \text{II}. \end{array} $$

We continue by applying the substitution \(\xi \mapsto 2^{j_{\text {sh}}} \xi \)

$$ \begin{array}{@{}rcl@{}} \text{II} &\lesssim& \sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} 2^{5/2j_{\text{sh}}} \left( {\int}_{\mathbb{R}^{2}} \frac{\min\{1, | 2^{j_{\text{sh}} - j_{\mathrm{w}}}{\xi}_{1}|^{{\alpha}_{\mathrm{w}}}\}}{\max\{1, |2^{j_{\text{sh}} - j_{\mathrm{w}}} {\xi}_{1}|^{{\beta}_{\mathrm{w}}^{\prime}}\}\max\{1, |2^{j_{\text{sh}} - j_{\mathrm{w}}}{\xi}_{2}|^{{\beta}_{\mathrm{w}}^{\prime}}\}} \right. \\ & &\quad \cdot \left. \frac{1}{\max\{1, |{\xi}_{1}|^{{\beta}_{\text{sh}}^{\prime}}\}\max\{1, |{\xi}_{2}|^{{\beta}_{\text{sh}}^{\prime}}\}} d\xi \right)^{2}. \\ &\lesssim &\sum\limits_{j_{\text{sh}} = 0}^{\infty} 2^{5/2j_{\text{sh}}} \left( {\int}_{\mathbb{R}^{2}} \frac{\min\{1, |2^{j_{\text{sh}} - j_{\mathrm{w}}} {\xi}_{1}|^{{\alpha}_{\mathrm{w}}}\}}{\max\{1, |{\xi}_{1}|^{{\beta}_{\text{sh}}^{\prime}}\}\max\{1, |{\xi}_{2}|^{{\beta}_{\text{sh}}^{\prime}}\}} d\xi \right)^{2} \\ &\lesssim &\sum\limits_{j_{\text{sh}} = 0}^{\infty} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-j_{\mathrm{w}})} \left( {\int}_{\mathbb{R}^{2}} \frac{ |{\xi}_{1}|^{{\alpha}_{\mathrm{w}}} }{\max\{1, |{\xi}_{1}|^{{\beta}_{\text{sh}}^{\prime}}\}\max\{1, |{\xi}_{2}|^{{\beta}_{\text{sh}}^{\prime}}\}} d\xi \right)^{2}. \end{array} $$

Since \({\beta }_{\text {sh}}^{\prime }-{\alpha }_{\mathrm {w}}>1\) we obtain that the integral above is finite and hence we conclude

$$ \begin{array}{@{}rcl@{}} \mathrm{I} \lesssim \sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-j_{\mathrm{w}})}. \end{array} $$
(10)

We rewrite the last sum above as

$$ \begin{array}{@{}rcl@{}} \sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-j_{\mathrm{w}})} = 2^{-2 {\alpha}_{\mathrm{w}} \epsilon j_{\mathrm{w}} }\sum\limits_{j_{\text{sh}} = 0}^{(2\tau)(j_{\mathrm{w}}-t)} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-(1-\epsilon)j_{\mathrm{w}})}. \end{array} $$
(11)

Since jw > 1/(2τ)jsh + t, we can now estimate

$$ \begin{array}{@{}rcl@{}} \sum\limits_{j_{\text{sh}} = 0}^{\infty} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-(1-\epsilon)j_{\mathrm{w}})} \lesssim 2^{-2 {\alpha}_{\mathrm{w}} (1-\epsilon)t} \sum\limits_{j_{\text{sh}}=0}^{\infty} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-(1-\epsilon) (1/(2\tau)j_{\text{sh}}) }. \end{array} $$

Since αw((1 − ε)/τ − 2) > 5/2 by assumption, the latter sum is finite. This leads to the estimate

$$ \begin{array}{@{}rcl@{}} \sum\limits_{j_{\text{sh}} = 0}^{\infty} 2^{5/2j_{\text{sh}} + 2{\alpha}_{\mathrm{w}}(j_{\text{sh}}-(1-\epsilon)j_{\mathrm{w}})} \lesssim 2^{-2 {\alpha}_{\mathrm{w}} (1-\epsilon)t}. \end{array} $$
(12)

Now (23) in combination with (22) and (21) implies together with (20) that

$$ \begin{array}{@{}rcl@{}} &&\sum\limits_{(j_{\text{sh}},k,m,\iota) \in {{\Lambda}_{0}^{c}}} \sum\limits_{(j_{\mathrm{w}},m^{\prime},1)\in {\Theta}_{\tau, t }^{c}} |\langle ({2^{-j_{\mathrm{w}} s}\omega_{j_{\mathrm{w}},m^{\prime},\upsilon}})^{d}, 2^{-j_{\text{sh}} s}{\psi}_{j_{\text{sh}},k,m,\iota} \rangle_{H^{s}({\Omega})}|^{2} \\ &\lesssim& \sum\limits_{j_{\mathrm{w}} = 0}^{\infty} 2^{-2 {\alpha}_{\mathrm{w}} \epsilon j_{\mathrm{w}} } 2^{-2 {\alpha}_{\mathrm{w}} (1-\epsilon)t}\lesssim 2^{-2 {\alpha}_{\mathrm{w}} (1-\epsilon)t}. \end{array} $$

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Grohs, P., Kutyniok, G., Ma, J. et al. Anisotropic multiscale systems on bounded domains. Adv Comput Math 46, 39 (2020). https://doi.org/10.1007/s10444-020-09784-0

Download citation

Keywords

  • Shearlets
  • Wavelets
  • Bounded domains
  • Sobolev spaces
  • Frame

Mathematics Subject Classification (2010)

  • 42C40
  • 65M60
  • 41A25
  • 65T99
  • 94A11