arXiv subjects

Rasmus Dalgas Kongskov

Publications and source records attributed to Rasmus Dalgas Kongskov.

3 recordsLinked to original sources

arXiv2017

Tomographic Reconstruction Methods for Decomposing Directional Components

Decomposition of tomographic reconstructions has many different practical application. We propose two new reconstruction methods that combines the task of tomographic reconstruction with object decomposition. We demonstrate these reconstruction methods in the context of decomposing directional objects into various directional components. Furthermore we propose a method for estimating the main direction in a directional object, directly from the measured computed tomography data. We demonstrate all the proposed methods on simulated and real samples to show their practical applicability. The numerical tests show that decomposition and reconstruction can combined to achieve a highly useful fibre-crack decomposition.

cs.CE↗

arXiv2017

Directional Total Generalized Variation Regularization

In inverse problems, prior information and a priori-based regularization techniques play important roles. In this paper, we focus on image restoration problems, especially on restoring images whose texture mainly follow one direction. In order to incorporate the directional information, we propose a new directional total generalized variation (DTGV) functional, which is based on total generalized variation (TGV) by Bredies \textit{et al}. [SIAM J. Imaging Sci., 3 (2010)]. After studying the mathematical properties of DTGV, we utilize it as regularizer and propose the L$^2$-DTGV variational model for solving image restoration problems. Due to the requirement of the directional information in DTGV, we give a direction estimation algorithm, and then apply a primal-dual algorithm to solve the minimization problem. Experimental results show the effectiveness of the proposed method for restoring the directional images. In comparison with isotropic regularizers like total variation and TGV, the improvement of texture preservation and noise removal is significant.

math.NA↗

arXiv2015

Noise Robustness of a Combined Phase Retrieval and Reconstruction Method for Phase-Contrast Tomography

Classical reconstruction methods for phase-contrast tomography consist of two stages: phase retrieval and tomographic reconstruction. A novel algebraic method combining the two was suggested by Kostenko et al. (Opt. Express, 21, 12185, 2013) and preliminary results demonstrating improved reconstruction compared to a two-stage method given. Using simulated free-space propagation experiments with a single sample-detector distance, we thoroughly compare the novel method with the two-stage method to address limitations of the preliminary results. We demonstrate that the novel method is substantially more robust towards noise; our simulations point to a possible reduction in counting times by an order of magnitude.

math.NA↗