Physics in Medicine and Biology · 2013 · 317 citations · 24 references
EngineeringBiomedical EngineeringRadiologyBiomedical ImagingDiagnostic ImagingDiagnostic RadiologyComputed TomographyComputed Tomography PhysicsImage CompressionImage ReconstructionMedical ImagingThree-dimensional ReconstructionCompressed SensingComputed Tomography ReconstructionCt Image ReconstructionData ConsistencyLimited-angle ReconstructionReconstruction AlgorithmsTotal Variation MinimizationX-ray Computed Tomography
This paper presents a compressed sensing (CS)-inspired reconstruction method for limited-angle computed tomography (CT). Currently, CS-inspired CT reconstructions are often performed by minimizing the total variation (TV) of a CT image subject to data consistency. A key to obtaining high image quality is to optimize the balance between TV-based smoothing and data fidelity. In the case of the limited-angle CT problem, the strength of data consistency is angularly varying. For example, given a parallel beam of x-rays, information extracted in the Fourier domain is mostly orthogonal to the direction of x-rays, while little is probed otherwise. However, the TV minimization process is isotropic, suggesting that it is unfit for limited-angle CT. Here we introduce an anisotropic TV minimization method to address this challenge. The advantage of our approach is demonstrated in numerical simulation with both phantom and real CT images, relative to the TV-based reconstruction.
24
The Split Bregman Method for L1-Regularized Problems
Tom Goldstein, Stanley Osher · SIAM Journal on Imaging Sciences · 2009 · 4.3K citations
The Mathematics of Computerized Tomography
Frank Natterer, Ge Wang · Medical Physics · 2002 · 2.6K citations · Full text
Super-resolution through Error Energy Reduction
R.W. Gerchberg · Optica Acta International Journal of Optics · 1974 · 988 citations
Computing, Computer Science, Physics +11