#### DMCA

## Compressive Imaging and Characterization of Sparse Light Deflection Maps (2014)

### Citations

7393 | Convex Optimization
- Boyd
- 2001
(Show Context)
Citation Context ... by ıC(z) = { 0 if z ∈ C, +∞ otherwise. (18) The convex indicator function is a proper, convex and lower-semicontinuous function and hence it satisfies the requirements of the Chambole-Pock algorithm =-=[7]-=-. Let B = {z ∈ RN | ‖y − z‖2 ≤ } be the `2 ball of radius , centred on the measurement vector y which is a convex set. Then the constraints ‖y −ΦSα‖2 ≤ and ‖y −ΦAs‖2 ≤ can be inserted into uncon... |

3580 | Compressed sensing
- Donoho
- 2006
(Show Context)
Citation Context ...vering full deflection spectra and also it uses binary modulation patterns to avoid SLM non-linearities. 4 Spread Spectrum Optical Compressive Sensing 4.1 Compressive sensing Compressive Sensing (CS) =-=[16, 10, 3, 30, 20]-=- is a new paradigm in signal sampling which envisages that any signal x = Ψα ∈ CN , where α is either sparse or compressible, can be tractably recovered from a few corrupted linear measurements of the... |

2253 | Nonlinear total variation based noise removal algorithms
- Rudin, Osher, et al.
- 1992
(Show Context)
Citation Context ... 5600 5000 (b) Figure 15: Example reconstructions of deflection spectrum for a diffractive MIOL. the deflection maps. The total-variation norm of a 2-dimensional discrete signal I(x, y) is defined as =-=[47, 43]-=- ‖I‖TV = ∑ x,y ‖∇I(x, y)‖2, (39) where ∇I(x, y) ∈ R2 is the gradient (horizontal and vertical) vector of the image at coordinates (x, y). The TV norm measures the amount of gradient in a signal, and h... |

1490 | Near optimal signal recovery from random projections: universal encoding strategies
- Candès, Tao
(Show Context)
Citation Context ...vering full deflection spectra and also it uses binary modulation patterns to avoid SLM non-linearities. 4 Spread Spectrum Optical Compressive Sensing 4.1 Compressive sensing Compressive Sensing (CS) =-=[16, 10, 3, 30, 20]-=- is a new paradigm in signal sampling which envisages that any signal x = Ψα ∈ CN , where α is either sparse or compressible, can be tractably recovered from a few corrupted linear measurements of the... |

694 | Model-based compressive sensing
- Baraniuk, Cevher, et al.
(Show Context)
Citation Context .... Indirectly probing a signal through its inner products, with known patterns, is a generalization of the classical sampling procedure where the modulation patterns are simply shifted delta functions =-=[3]-=-. Therefore, inner products with known modulation patterns are generalized samples of the signal. From now on, we will simply refer to these inner product samples as measurements. In the context of sa... |

533 | Sparse MRI: The Application of Compressed Sensing for Rapid
- Lustig, Donoho, et al.
- 2007
(Show Context)
Citation Context ...plementation for optical systems [55], has found several applications in imaging, beginning with the famous single pixel camera [17] to recent applications in, for example, magnetic resonance imaging =-=[31, 32, 38]-=-, astronomical imaging [6], radio interferometry [54], hyper spectral imaging [22] and biological imaging [48]. Contributions and organization of the article: The contributions of this article are two... |

433 | T.: A first-order primal-dual algorithm for convex problems with applications to imaging
- Chambolle, Pock
- 2011
(Show Context)
Citation Context ... arg min s ‖Ψ∗s‖1 subject to ‖y −ΦAs‖2 ≤ and s ∈ RN+ . (13) 6 Numerical method To solve the problems Eqs. (12) and (13), we make use of a primal-dual method called the Chambolle-Pock (CP) algorithm =-=[11]-=-. Chambolle-Pock algorithm solves primal-dual forms of unconstrained convex problems and it relies on proximal operators of the functions involved in the objective. It has a flexible structure, which ... |

347 | Introduction to the non-asymptotic analysis of random matrices
- Vershynin
- 2012
(Show Context)
Citation Context ...he form of a systematic bias in the reconstructed spectrum, which is tolerable for our purposes. 4A similar argument could be given if the noise is assumed to be sub-Gaussian, e.g., for bounded noise =-=[53]-=-. 18 7.2.2 Input SNR We also define a notion of input SNR using the measurements made without any test object. In the absence of any signal noise and measurement noise, the norm of the measurement vec... |

295 | Single-pixel imaging via compressive sampling
- Duarte, Davenport, et al.
(Show Context)
Citation Context ...ng deflection spectra. Compressed sensing, despite some difficulty in implementation for optical systems [55], has found several applications in imaging, beginning with the famous single pixel camera =-=[17]-=- to recent applications in, for example, magnetic resonance imaging [31, 32, 38], astronomical imaging [6], radio interferometry [54], hyper spectral imaging [22] and biological imaging [48]. Contribu... |

278 |
Combettes, Convex Analysis and Monotone Operator Theory in Hilbert Spaces
- Bauschke, L
- 2010
(Show Context)
Citation Context ...nction. 6.1 Chambolle-Pock algorithm Let K : RN → RM be a continuous linear operator with a bounded norm. Let F : RM → [0,+∞] and G : RN → [0,+∞] be two proper, convex, lower-semicontinuous functions =-=[4]-=-. The ChambollePock algorithm is used to solve saddle-point problems of the form min u∈RN max v∈RM 〈Ku,v〉+G(u)− F ∗(v). (14) It can be seen that Eq. (14) is the primal-dual formulation of the primal m... |

253 |
Sparse and Redundant Representations - From Theory to
- Elad
- 2010
(Show Context)
Citation Context ...ling theorem. Alternatively, if the underlying signal can be expressed or approximated by a linear combination of a few basis vectors, then results in sparse signal recovery and in compressed sensing =-=[19, 20]-=- show that with appropriate (random) patterns, one can successfully recover such a signal from a fewer number of measurements compared to the dimension of its ambient domain. This signal recovery proc... |

221 |
A Wavelet Tour of Signal Processing: The Sparse Way
- Mallat
- 2009
(Show Context)
Citation Context ...parse representationa of deflection spectra. One choice is to use wavelets as they are suitable for signals which are piecewise regular. In our work, we use Daubechies 16 tap orthogonal wavelet basis =-=[34]-=-, due to the expected smooth and non-dispersive shapes of deflection spectra. Representation in a wavelet basis is translation variant and hence the sparsity of the representation might be affected by... |

156 | Compressed sensing with coherent and redundant dictionaries
- Candès, Eldar, et al.
- 2010
(Show Context)
Citation Context ...., the usual wavelet transform without the decimation [34]. The non-uniqueness of the decomposition of a signal in such a redundant frame allows us to represent signals in their sparsest possible way =-=[19, 8]-=-. Our work also study the advantage of using a UDWT built on the Daubechies 16 tap filter in all our reconstruction experiments. 5.2 Synthesis and analysis formulations The solution of the optimizatio... |

151 | Compressive Sensing and Structured Random Matrices
- Rauhut
- 2010
(Show Context)
Citation Context ..., satisfies ‖α− α̂‖2 = O (‖α−αK‖1√ K + ) (6) with a probability at least 1 − N−γ log3(N), where αK is the best K-term approximation to the vector α, ‖α−αK‖2 ≤ ‖α−α′‖2 for all α′ such that ‖α′‖0 ≤ K =-=[9, 40]-=-. Similar type of result also holds when the measurement matrix is derived from an orthonormal basis Γ ∈ CN×N . In such a case, a M ×N measurement matrix is of the form Γ∗Ω, the conjugate 2A matrix wh... |

142 | Analysis versus synthesis in signal priors
- Elad, Milanfar, et al.
- 2007
(Show Context)
Citation Context ...for a solution directly in the signal space, which, when analyzed with Ψ, has a sparse representation and also 10 agrees with the measurements. Hence, this approach is called as the analysis approach =-=[18, 36]-=-. In our work, we solve both the synthesis and analysis problems with a UDWT dictionary. To summarize, our work will analyze the benefit of using one of the following schemes for reconstructing sparse... |

112 | Compressive sensing by random convolution
- Romberg
(Show Context)
Citation Context ...e transforms. On a related note, some researchers have investigated other ways of designing structured random sensing matrices. Notable of them are the ones based on convolution with random sequences =-=[25, 51, 42, 56]-=-, where the signal is convolved with a random sequence before subsampling. Even this scheme is shown to be universal and works well with any sparsity basis [42]. Convolutions can be 8 implemented as m... |

101 | Signal processing with compressive measurements
- Davenport, Boufounos, et al.
- 2010
(Show Context)
Citation Context ...formation about the deflection spectra, sufficient to characterize the shapes of objects under consideration. This work is on the lines of compressed domain signal processing and parameter estimation =-=[14, 13, 21]-=-. We develop a simplified description of deflection spectrum, which is characterized by a pair of translation parameters. This is inspired by the fact that when the surface of an object is smooth, the... |

94 |
Random filters for compressive sampling and reconstruction
- Tropp, Wakin, et al.
- 2006
(Show Context)
Citation Context ...e transforms. On a related note, some researchers have investigated other ways of designing structured random sensing matrices. Notable of them are the ones based on convolution with random sequences =-=[25, 51, 42, 56]-=-, where the signal is convolved with a random sequence before subsampling. Even this scheme is shown to be universal and works well with any sparsity basis [42]. Convolutions can be 8 implemented as m... |

92 | Toeplitz compressed sensing matrices with applications to sparse channel estimation
- Haupt, Bajwa, et al.
- 2010
(Show Context)
Citation Context ...e transforms. On a related note, some researchers have investigated other ways of designing structured random sensing matrices. Notable of them are the ones based on convolution with random sequences =-=[25, 51, 42, 56]-=-, where the signal is convolved with a random sequence before subsampling. Even this scheme is shown to be universal and works well with any sparsity basis [42]. Convolutions can be 8 implemented as m... |

71 | The smashed filter for compressive classification and target recognition,” in
- Davenport, Duarte, et al.
- 2007
(Show Context)
Citation Context ...formation about the deflection spectra, sufficient to characterize the shapes of objects under consideration. This work is on the lines of compressed domain signal processing and parameter estimation =-=[14, 13, 21]-=-. We develop a simplified description of deflection spectrum, which is characterized by a pair of translation parameters. This is inspired by the fact that when the surface of an object is smooth, the... |

51 |
Schlieren and shadowgraph techniques. In: Visualizing Phenomena in Transparent Media
- Settles
- 2001
(Show Context)
Citation Context ...e3 for PlanosConvex tan θϕ Figure 1: Left, illustration of a deflection spectrum. Right, a typical (projected) deflection spectrum sp for a plano convex lens of optical power 25.12D. light deflection =-=[45]-=-. These schlieren deflectometers operate by converting deflections into grayscale values and can be employed for any medium (solid, liquid and gases). Applications of schlieren techniques include flow... |

48 |
Sparsity and incoherence in compressive sampling,” Inverse problems
- Candes, Romberg
- 2007
(Show Context)
Citation Context ..., satisfies ‖α− α̂‖2 = O (‖α−αK‖1√ K + ) (6) with a probability at least 1 − N−γ log3(N), where αK is the best K-term approximation to the vector α, ‖α−αK‖2 ≤ ‖α−α′‖2 for all α′ such that ‖α′‖0 ≤ K =-=[9, 40]-=-. Similar type of result also holds when the measurement matrix is derived from an orthonormal basis Γ ∈ CN×N . In such a case, a M ×N measurement matrix is of the form Γ∗Ω, the conjugate 2A matrix wh... |

44 |
Digital phase-shifting interferometry: a simple error-compensating phase calculation algorithm,’’
- Hariharan, Oreb, et al.
- 1987
(Show Context)
Citation Context ...soids are made to obtain extra information to solve for the unknowns in Eq. (3). Then deviation angles are numerically decoded and 2pi phase unwrapped from the measurements yk using n-step algorithms =-=[24, 2]-=-. PSS is a simple and effective method for measuring deflection angles and has been successfully used in tomographic applications such as refractive index map reconstruction [5, 23]. However, this met... |

42 | Compressed sensing in astronomy,”
- Bobin, Starck, et al.
- 2008
(Show Context)
Citation Context ...5], has found several applications in imaging, beginning with the famous single pixel camera [17] to recent applications in, for example, magnetic resonance imaging [31, 32, 38], astronomical imaging =-=[6]-=-, radio interferometry [54], hyper spectral imaging [22] and biological imaging [48]. Contributions and organization of the article: The contributions of this article are twofold. Firstly, we demonstr... |

35 |
MAGNOR M.: Image-Based Tomographic Reconstruction of Flames
- IHRKE
(Show Context)
Citation Context ... operate by converting deflections into grayscale values and can be employed for any medium (solid, liquid and gases). Applications of schlieren techniques include flow modeling and computer graphics =-=[15, 46, 26]-=-. In this article, we consider an instance of schlieren deflectometer, which is used for characterizing solid transparent objects such as optical lenses. Most of the alternative optical modalities for... |

30 | Compressed sensing imaging techniques for radio interferometry
- Wiaux, Jacques, et al.
(Show Context)
Citation Context ...ications in imaging, beginning with the famous single pixel camera [17] to recent applications in, for example, magnetic resonance imaging [31, 32, 38], astronomical imaging [6], radio interferometry =-=[54]-=-, hyper spectral imaging [22] and biological imaging [48]. Contributions and organization of the article: The contributions of this article are twofold. Firstly, we demonstrate a novel way to compress... |

27 | Y.Wiaux, “Universal and efficient compressed sensing by spread spectrum and application to realistic Fourier imaging techniques,”EURASIP
- Puy, Vandergheynst, et al.
(Show Context)
Citation Context ...nd present a numerical method for their reconstruction along with experimental results. To this end, we present a design of optical 3 modulation patterns based on spread spectrum1 compressive sensing =-=[39]-=-. This framework not only enables us to leverage the power of random measurements, as advocated by compressed sensing theory, but also makes the numerical methods exploit the advantage of fast algorit... |

20 |
Compressed sensing for practical optical imaging systems: a tutorial
- Willett, Marcia, et al.
(Show Context)
Citation Context ...dulation patterns. This is a tailor-made situation for adopting compressed sensing for recovering deflection spectra. Compressed sensing, despite some difficulty in implementation for optical systems =-=[55]-=-, has found several applications in imaging, beginning with the famous single pixel camera [17] to recent applications in, for example, magnetic resonance imaging [31, 32, 38], astronomical imaging [6... |

19 |
Foucart and Holger Rauhut. A Mathematical Introduction to Compressive Sensing. Applied and Numerical Harmonic Analysis. Birkhäuser
- Simon
- 2013
(Show Context)
Citation Context ...ling theorem. Alternatively, if the underlying signal can be expressed or approximated by a linear combination of a few basis vectors, then results in sparse signal recovery and in compressed sensing =-=[19, 20]-=- show that with appropriate (random) patterns, one can successfully recover such a signal from a fewer number of measurements compared to the dimension of its ambient domain. This signal recovery proc... |

19 |
Practical compressive sensing with Toeplitz and circulant matrices. Rice University CAAM
- Yin, Morgan, et al.
- 2010
(Show Context)
Citation Context |

16 | Compressive sampling of pulse trains: spread the spectrum
- Naini, Gribonval, et al.
- 2009
(Show Context)
Citation Context ...design of the sensing matrix Φ in order to maximize the information captured in each sample. To this end, we design the modulation patterns relying on the theory of spread spectrum compressed sensing =-=[35, 39]-=-. As the forward measurement model in Eq. (1) and the recovery problem formulation (developed in future sections) are the same for all the locations of the CCD, we suspend the use of the subscript k t... |

16 |
Proximal algorithms. Foundations and Trends in Optimization
- Parikh, Boyd
- 2013
(Show Context)
Citation Context ...n z∈RN 1 2γ ‖u− z‖22 + F (z). (17) It can be shown that the proximal operator generalizes a simple gradient descent of F , in the form an implicit subgradient descent, even if F is not differentiable =-=[37]-=-. Moreover, proximal operators of several functions commonly used in signal processing have closed forms which are easy to evaluate, e.g., the proximal operator of `1 norm is soft thresholding [12, 37... |

14 | Spread spectrum Magnetic Resonance Imaging
- Puy, Marques, et al.
- 2012
(Show Context)
Citation Context ...plementation for optical systems [55], has found several applications in imaging, beginning with the famous single pixel camera [17] to recent applications in, for example, magnetic resonance imaging =-=[31, 32, 38]-=-, astronomical imaging [6], radio interferometry [54], hyper spectral imaging [22] and biological imaging [48]. Contributions and organization of the article: The contributions of this article are two... |

12 |
Combettes and Jean-Christophe Pesquet. Proximal splitting methods in signal processing
- Patrick
- 2011
(Show Context)
Citation Context ...le [37]. Moreover, proximal operators of several functions commonly used in signal processing have closed forms which are easy to evaluate, e.g., the proximal operator of `1 norm is soft thresholding =-=[12, 37]-=-. Further, the proximal operator of the conjugate function F ∗ is easily derived using Moreau’s identity [11]. The proximal operators for the functions appearing in the minimization problems Eq. (5) a... |

8 | Compressive source separation: Theory and methods for hyperspectral imaging,” arxiv:1208.4505,
- Golbabaee, Arberet, et al.
- 2012
(Show Context)
Citation Context ...g with the famous single pixel camera [17] to recent applications in, for example, magnetic resonance imaging [31, 32, 38], astronomical imaging [6], radio interferometry [54], hyper spectral imaging =-=[22]-=- and biological imaging [48]. Contributions and organization of the article: The contributions of this article are twofold. Firstly, we demonstrate a novel way to compressively acquire sparse deflecti... |

6 | Compressive Optical Deflectometric Tomography: A Constrained Total-Variation Minimization Approach
- Gonzalez, Jacque, et al.
- 2014
(Show Context)
Citation Context ... n-step algorithms [24, 2]. PSS is a simple and effective method for measuring deflection angles and has been successfully used in tomographic applications such as refractive index map reconstruction =-=[5, 23]-=-. However, this method is not stable for deflection spectra that are spread out, unlike a point or even of estimating several main deflection angles for each object location. Therefore, the richness o... |

5 | Compressive parameter estimation for sparse translationinvariant signals using polar interpolation,” submitted for publication
- Fyhn, Duarte, et al.
(Show Context)
Citation Context ...formation about the deflection spectra, sufficient to characterize the shapes of objects under consideration. This work is on the lines of compressed domain signal processing and parameter estimation =-=[14, 13, 21]-=-. We develop a simplified description of deflection spectrum, which is characterized by a pair of translation parameters. This is inspired by the fact that when the surface of an object is smooth, the... |

4 |
Optical Shop Testing (Wiley
- Malacara
- 2007
(Show Context)
Citation Context ...meter, which is used for characterizing solid transparent objects such as optical lenses. Most of the alternative optical modalities for characterizing transparent objects are based on interferometry =-=[33]-=-. However, techniques based on optical interferometry are very sensitive to vibrations and also need precise calibrations for the methods to work successfully. On the other hand, thanks to the nature ... |

4 | Analysis `1-recovery with frames and Gaussian measurements
- Kabanava, Rauhut
- 2013
(Show Context)
Citation Context ... . (11) In our work, we make use of tight frames based sparse representation for analysis based recovery. Recovery guarantees when Ψ is not a tight frame and measurements are Gaussian can be found in =-=[41]-=-. At the outset, the two problems, Eqs. (5) and (10), look identical, except for the placement of the matrix Ψ. In fact, when Ψ is an orthonormal basis the location of the sparsity basis has no bearin... |

4 |
Colour-coding schlieren techniques for the optical study of heat and fluid flow
- Settles
- 1985
(Show Context)
Citation Context ... operate by converting deflections into grayscale values and can be employed for any medium (solid, liquid and gases). Applications of schlieren techniques include flow modeling and computer graphics =-=[15, 46, 26]-=-. In this article, we consider an instance of schlieren deflectometer, which is used for characterizing solid transparent objects such as optical lenses. Most of the alternative optical modalities for... |

3 |
Optical deflection tomography with the phase-shifting schlieren
- Beghuin, Dewandel, et al.
(Show Context)
Citation Context ... n-step algorithms [24, 2]. PSS is a simple and effective method for measuring deflection angles and has been successfully used in tomographic applications such as refractive index map reconstruction =-=[5, 23]-=-. However, this method is not stable for deflection spectra that are spread out, unlike a point or even of estimating several main deflection angles for each object location. Therefore, the richness o... |

3 |
Phase-shifting schlieren: high-resolution quantitative schlieren that uses the phase-shifting technique principle
- Joannes, Dubois, et al.
(Show Context)
Citation Context ...he purpose of emphasizing the advantage of the compressive sensing method, we shall briefly explain the existing method. For a detailed description of the same, the readers are encouraged to refer to =-=[28, 27]-=-. The system is configured to work using several phase shifted sinusoidal patterns (for modulation) in order to measure deflections and this configuration is named Phase Shifting Schlieren (PSS). In P... |

3 |
Michael Elad, and Rémi Gribonval, The Cosparse Analysis Model and Algorithms
- Nam, Davies
(Show Context)
Citation Context ...for a solution directly in the signal space, which, when analyzed with Ψ, has a sparse representation and also 10 agrees with the measurements. Hence, this approach is called as the analysis approach =-=[18, 36]-=-. In our work, we solve both the synthesis and analysis problems with a UDWT dictionary. To summarize, our work will analyze the benefit of using one of the following schemes for reconstructing sparse... |

3 |
Goniophotometry: new calibration method and instrument design
- Sauter
- 1995
(Show Context)
Citation Context ...gible. Measuring deflection angles in a straightforward way using goniophotometer is a cumbersome and elaborate process, and it is only suited for large dynamic of deflection angles and high contrast =-=[44]-=-. However, in the present context, the deflection information is observed using an indirect method. The particular optical setup at our disposal, which will be described in Sec. 2, measures the spectr... |

2 |
Jérome Bobin, Makhlad Chahid, Hamed Shams Mousavi, Emmanuel Candes, and Maxime Dahan. Compressive fluorescence microscopy for biological and hyperspectral imaging
- Studer
(Show Context)
Citation Context ...el camera [17] to recent applications in, for example, magnetic resonance imaging [31, 32, 38], astronomical imaging [6], radio interferometry [54], hyper spectral imaging [22] and biological imaging =-=[48]-=-. Contributions and organization of the article: The contributions of this article are twofold. Firstly, we demonstrate a novel way to compressively acquire sparse deflection spectra in a schlieren de... |

1 |
Two-step phase-shifting algorithm
- Almazán-Cuéllar, Malacara-Hernández
(Show Context)
Citation Context ...soids are made to obtain extra information to solve for the unknowns in Eq. (3). Then deviation angles are numerically decoded and 2pi phase unwrapped from the measurements yk using n-step algorithms =-=[24, 2]-=-. PSS is a simple and effective method for measuring deflection angles and has been successfully used in tomographic applications such as refractive index map reconstruction [5, 23]. However, this met... |

1 |
Schlieren photography a short bibliography and review
- Davies
- 1981
(Show Context)
Citation Context ... operate by converting deflections into grayscale values and can be employed for any medium (solid, liquid and gases). Applications of schlieren techniques include flow modeling and computer graphics =-=[15, 46, 26]-=-. In this article, we consider an instance of schlieren deflectometer, which is used for characterizing solid transparent objects such as optical lenses. Most of the alternative optical modalities for... |

1 |
High-resolution shape measurements with phase-shifting schlieren (PSS
- Joannes, Beghuin, et al.
- 2004
(Show Context)
Citation Context ...he purpose of emphasizing the advantage of the compressive sensing method, we shall briefly explain the existing method. For a detailed description of the same, the readers are encouraged to refer to =-=[28, 27]-=-. The system is configured to work using several phase shifted sinusoidal patterns (for modulation) in order to measure deflections and this configuration is named Phase Shifting Schlieren (PSS). In P... |

1 |
Understanding the diffractive bifocal contact lens
- Klein
- 1993
(Show Context)
Citation Context ...lens (s) f f p yp Tele-centric system(TS) θp O r r = f tan θp −d d (a) (b) Figure 2: (a) 2D schematic of schlieren deflectometer and (b) a commercially available NIMOTMschlieren deflectometer. lenses =-=[29]-=-, i.e., lenses displaying multiple foci due to their special design of surfaces. 2 Optical setup and Notations The schematic of the schlieren deflectometer used in our work to measure deflection spect... |

1 |
Compressed sensing: Theory and applications. Arxiv preprint arXiv:1203.3815
- Kutyniok
- 2012
(Show Context)
Citation Context ...vering full deflection spectra and also it uses binary modulation patterns to avoid SLM non-linearities. 4 Spread Spectrum Optical Compressive Sensing 4.1 Compressive sensing Compressive Sensing (CS) =-=[16, 10, 3, 30, 20]-=- is a new paradigm in signal sampling which envisages that any signal x = Ψα ∈ CN , where α is either sparse or compressible, can be tractably recovered from a few corrupted linear measurements of the... |

1 |
Compressive acquisition of sparse deflectometric maps
- Sudhakar, Jacques, et al.
- 2013
(Show Context)
Citation Context ...c imaging in reducing the number of measurements. A description of the optical setup and a basic reconstruction method, along with the results, were briefly presented in our conference communications =-=[50, 49]-=-. However, the present article contains several new material including the details of modulation pattern design and also new strategies for reconstruction, along with new contributions on how to use d... |

1 | Compressive schlieren deflectometry
- Sudhakar, Jacques, et al.
- 2013
(Show Context)
Citation Context ...c imaging in reducing the number of measurements. A description of the optical setup and a basic reconstruction method, along with the results, were briefly presented in our conference communications =-=[50, 49]-=-. However, the present article contains several new material including the details of modulation pattern design and also new strategies for reconstruction, along with new contributions on how to use d... |