Mask iterative hard thresholding algorithms for sparse image reconstruction of objects with known contour

@article{Dogandzic2011MaskIH,
  title={Mask iterative hard thresholding algorithms for sparse image reconstruction of objects with known contour},
  author={Aleksandar Dogandzic and Renliang Gu and Kun Qiu},
  journal={2011 Conference Record of the Forty Fifth Asilomar Conference on Signals, Systems and Computers (ASILOMAR)},
  year={2011},
  pages={2111-2116}
}
We develop mask iterative hard thresholding algorithms (mask IHT and mask DORE) for sparse image reconstruction of objects with known contour. The measurements follow a noisy underdetermined linear model common in the compressive sampling literature. Assuming that the contour of the object that we wish to reconstruct is known and that the signal outside the contour is zero, we formulate a constrained residual squared error minimization problem that incorporates both the geometric information (i… 

Figures from this paper

Algorithms for sparse X-ray CT image reconstruction of objects with known contour
TLDR
A constrained residual squared error minimization criterion is proposed that incorporates both the knowledge of the object's contour and signal sparsity in an appropriate transform domain and convex relaxation and greedy approaches to approximately solving this minimization problem are presented.
Sparse Signal Reconstruction via ECME Hard Thresholding
TLDR
A probabilistic model for sparse signal reconstruction is proposed and several novel algorithms for computing the maximum likelihood (ML) parameter estimates under this model are developed.
Blind beam-hardening correction from Poisson measurements
We develop a sparse image reconstruction method for Poisson-distributed polychromatic X-ray computed tomography (CT) measurements under the blind scenario where the material of the inspected object
A fast proximal gradient algorithm for reconstructing nonnegative signals with sparse transform coefficients
TLDR
This work develops a fast proximal gradient scheme for reconstructing nonnegative signals that are sparse in a transform domain from underdetermined measurements and demonstrates that, by exploiting both the nonnegativity of the underlying signal and sparsity of its wavelet coefficients, it can achieve significantly better reconstruction performance than the existing methods.
Reconstruction of Nonnegative Sparse Signals Using Accelerated Proximal-Gradient Algorithms
TLDR
An accelerated proximal-gradient scheme for reconstructing nonnegative signals that are sparse in a transform domain from underdetermined measurements and can achieve significantly better reconstruction performance than the existing methods is developed.
Convex-Set–Constrained Sparse Signal Recovery: Theory and Applications
TLDR
This work is the first to upper-bound the regularization constant for total-variation (TV) regularizations for convex-set–constrained sparse signal recovery problem behind the PNPG framework.
Sparse X-ray CT image reconstruction and blind beam hardening correction via mass attenuation discretization
  • Renliang Gu, Aleksandar Dogandzic
  • Computer Science
    2013 5th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)
  • 2013
TLDR
A nonlinear sparse X-ray computed tomography (CT) image reconstruction method that accounts for beam hardening effects due to polychromatic X-rays sources and compares the proposed method with the existing approaches, which ignore the poly chromatic nature of the measurements or sparsity of the density map image.
Reconstructing Structured Sparse Signals from Compressive Samples via a Max-Product EM Algorithm
TLDR
This work proposes a Bayesian expectation-maximization (EM) algorithm for reconstructing structured approximately sparse signals via belief propagation that achieves better reconstruction performance compared with the state-of-the-art methods.
Polychromatic sparse image reconstruction and mass attenuation spectrum estimation via B-spline basis function expansion
TLDR
A block coordinate-descent algorithm for constrained minimization of a penalized negative log-likelihood function, where constraints and penalty terms ensure nonnegativity of the spline coefficients and sparsity of the density map image in the wavelet domain is developed.
Projected Nesterov's Proximal-Gradient Algorithm for Sparse Signal Recovery
TLDR
An integrated derivation of the momentum acceleration and proofs of the objective function convergence rate and convergence of the iterates, which account for adaptive step size, inexactness of the Iterative proximal mapping, and the convex-set constraint are presented.
...
1
2
...

References

SHOWING 1-10 OF 13 REFERENCES
Algorithms for sparse X-ray CT image reconstruction of objects with known contour
TLDR
A constrained residual squared error minimization criterion is proposed that incorporates both the knowledge of the object's contour and signal sparsity in an appropriate transform domain and convex relaxation and greedy approaches to approximately solving this minimization problem are presented.
Gradient Projection for Sparse Reconstruction: Application to Compressed Sensing and Other Inverse Problems
TLDR
This paper proposes gradient projection algorithms for the bound-constrained quadratic programming (BCQP) formulation of these problems and test variants of this approach that select the line search parameters in different ways, including techniques based on the Barzilai-Borwein method.
A Fast Algorithm for Sparse Reconstruction Based on Shrinkage, Subspace Optimization, and Continuation
TLDR
The code FPC_AS embeds this basic two-stage algorithm in a continuation (homotopy) approach by assigning a decreasing sequence of values to $\mu$ and exhibits state-of-the-art performance in terms of both its speed and its ability to recover sparse signals.
Compressive sensing of images with a priori known spatial support
TLDR
In this work, incorporation of either complete or partial a priori knowledge of object spatial support into the compressive sensing (CS) framework is investigated and the proposed augmented reconstruction model was shown to be robust to inaccuracies in the estimated object support.
Iterative Hard Thresholding for Compressed Sensing
Computational Methods for Sparse Solution of Linear Inverse Problems
TLDR
This paper surveys the major practical algorithms for sparse approximation with specific attention to computational issues, to the circumstances in which individual methods tend to perform well, and to the theoretical guarantees available.
Principles of Computerized Tomographic Imaging
The total attenuation suffered by one beam of x-rays as it travels in a straight line through an object can be represented by a line integral. Combining a set of line integrals forms a projection.
Matrix Algebra From a Statistician's Perspective
Preface. - Matrices. - Submatrices and partitioned matricies. - Linear dependence and independence. - Linear spaces: row and column spaces. - Trace of a (square) matrix. - Geometrical considerations.
A fast algorith m for sparse reconstruction based on shrinkage, subspace optimi zation, and continuation,”SIAM
  • J. Sci. Comput. , vol. 32,
  • 2010
Computational methods for s parse solution of linear inverse problems
  • Proc. IEEE, vol. 98, no. 6, pp. 948–958, 2010.
  • 2010
...
1
2
...