The Ettention software package

Dahmen, Tim; Marsalek, Lukas; Marniok, Nico; Turoňová, Beata; Bogachev, Sviatoslav; Trampert, Patrick; Nickels, Stefan; Slusallek, Philipp

doi:10.1016/j.ultramic.2015.10.012

Cited by 17 publications

(15 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our implementation was based on the framework Ettention [19], which is a software package for tomographic reconstructions in electron tomography running on high-performance computing devices.…”

Section: Methodsmentioning

confidence: 99%

“…1). Tomographic reconstruction was performed with the software Ettention [19] using a volume resolution of 2048 2 × 192 voxels and an isotropic voxel size of 20 m. For this purpose, the SART was implemented as described above using blobs as well as voxels as basis functions. The relaxation factor λ was set constantly to 0.3 for all reconstructions, as finding optimal relaxation factors was not a goal of this study.…”

Section: Evaluation Methodsmentioning

confidence: 99%

“…This lead to the hypothesis that incorporating blob basis functions into a state of the art reconstruction algorithm can enhance reconstruction quality and resolution. To validate this hypothesis, we implemented voxel based and blob based SART for CL within the framework Ettention [19] to be able to compare reconstruction capabilities within one unified codebase.…”

Section: P Trampert Et Al / Spherically Symmetric Volume Elements Amentioning

confidence: 99%

See 2 more Smart Citations

Spherically symmetric volume elements as basis functions for image reconstructions in computed laminography

Trampert

Vogelgesang

Schorr

et al. 2017

Journal of X-Ray Science and Technology: Clinical Applications

Self Cite

View full text Add to dashboard Cite

Abstract. BACKGROUND:Laminography is a tomographic technique that allows three-dimensional imaging of flat, elongated objects that stretch beyond the extent of a reconstruction volume. Laminography datasets can be reconstructed using iterative algorithms based on the Kaczmarz method. OBJECTIVE: The goal of this study is to develop a reconstruction algorithm that provides superior reconstruction quality for a challenging class of problems. METHODS: Images are represented in computer memory using coefficients over basis functions, typically piecewise constant functions (voxels). By replacing voxels with spherically symmetric volume elements (blobs) based on generalized KaiserBessel window functions, we obtained an adapted version of the algebraic reconstruction technique. RESULTS: Band-limiting properties of blob functions are beneficial particular in the case of noisy projections and if only a limited number of projections is available. In this case, using blob basis functions improved the full-width-at-half-maximum resolution from 10.2 ± 1.0 to 9.9 ± 0.9 (p value = 2.3·10 −4 ). For the same dataset, the signal-to-noise ratio improved from 16.1 to 31.0. The increased computational demand per iteration is compensated for by a faster convergence rate, such that the overall performance is approximately identical for blobs and voxels. CONCLUSIONS: Despite the higher complexity, tomographic reconstruction from computed laminography data should be implemented using blob basis functions, especially if noisy data is expected.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Evaluation Methodsmentioning

confidence: 99%

Section: P Trampert Et Al / Spherically Symmetric Volume Elements Amentioning

confidence: 99%

See 1 more Smart Citation

Spherically symmetric volume elements as basis functions for image reconstructions in computed laminography

Trampert

Vogelgesang

Schorr

et al. 2017

Journal of X-Ray Science and Technology: Clinical Applications

Self Cite

View full text Add to dashboard Cite

show abstract

“…In real space blobs have a compact support, that means blobs enforce band-limiting properties and radial symmetry. We implemented SART based on blobs and based on voxels in the framework Ettention [3]. The different geometric setup of CL compared to standard Computed Tomography setups was adapted in the SART formulation, so that forward and backward projections could be implemented easily for iteratively reconstructing the three dimensional volume.To evaluate the new method, a welded aluminum piece with a crack was recorded using the CLARA geometry.…”

mentioning

confidence: 99%

Blob-based Algebraic Reconstruction Technique for Computed Laminography

2018

Self Cite

View full text Add to dashboard Cite

Computed Laminography (CL) is an X-Ray based technique for acquiring flat samples, meaning one dimension of a sample is much smaller than the others. Tailored to this technique the CLARA geometry setup [1] is used to gather projections of a specimen. These projections are processed by an iterative algorithm based on the Kaczmarz method to obtain a three dimensional dataset of the specimen. This class of algorithms suffers from artifacts, especially for datasets with a poor signal-to-noise ratio (SNR) and when only a limited number of projections is available. We proposed a new method based on the simultaneous algebraic reconstruction technique (SART), where voxel basis functions are replaced with blob basis functions that have band-limiting properties to suppress artifacts [2].Blobs are spherically symmetric volume elements that have a ball shape as opposed to voxels, which have a cubic shape. In real space blobs have a compact support, that means blobs enforce band-limiting properties and radial symmetry. We implemented SART based on blobs and based on voxels in the framework Ettention [3]. The different geometric setup of CL compared to standard Computed Tomography setups was adapted in the SART formulation, so that forward and backward projections could be implemented easily for iteratively reconstructing the three dimensional volume.To evaluate the new method, a welded aluminum piece with a crack was recorded using the CLARA geometry. Two datasets were acquired, projections with low noise and a SNR of 9.0, and projections with high noise and a SNR of 6.12. Reconstructions were performed based on 100 and based on 400 projections for both sets. The four different setups were compared for voxel based and for blob based reconstructions. Besides visual inspection two statistical figures of merit were computed for evaluation. For assessment of the reconstruction quality the full width at half maximum (FWHM) of the point spread function and the slope of edge profiles were measured at different positions in the reconstructed volume.For each setup we determined the optimal reconstruction experimentally. The best results of each setup were then compared. The reconstructions of low noise data were similar in quality with blobs and voxels. However, the high noise data reconstructions for blobs had a significantly higher resolution than those with voxels, which was mirrored by p-values from a two-tailed t-test below 2.3•10 -4 for FWHM and below 2.2•10 -16 for slopes. A visual comparison (Figure 1) confirmed this result.The presented blob based reconstruction scheme for laminographic datasets is superior to normal voxel based approaches, especially when a high level of noise is present and with a small number of input projections that does not fulfill the Crowther criterion. The superior reconstruction quality was assessed quantitatively using FWHM, slopes of edge profiles, and SNR. Consequently, it is recommendable to use blobs as basis function for laminography data reconstructions. Blobs deliver equal or better results ...

show abstract

“…The new algorithm was implemented with the software package Ettention [2] and applied to the datasets with fixed parameters. We assessed the capabilities of the method with several evaluation metrics, namely root mean square error (RMSE), Peak signal-to-noise ratio (PSNR), and the structural similarity index (SSIM).…”

mentioning

confidence: 99%