TY - JOUR A1 - Tsamos, Athanasios A1 - Evsevleev, Sergei A1 - Fioresi, Rita A1 - Faglioni, Francesco A1 - Bruno, Giovanni T1 - A Complete Strategy to Achieve High Precision Automatic Segmentation of Challenging Experimental X‐Ray Computed Tomography Data Using Low‐Resemblance Synthetic Training Data N2 - It is shown that preconditioning of experimental X‐ray computed tomography (XCT) data is critical to achieve high‐precision segmentation scores. The challenging experimental XCT datasets and deep convolutional neural networks (DCNNs) are used that are trained with low‐resemblance synthetic XCT data. The material used is a 6‐phase Al–Si metal matrix composite‐reinforced with ceramic fibers and particles. To achieve generalization, in our past studies, specific data augmentation techniques were proposed for the synthetic XCT training data. In addition, two toolsets are devised: (1) special 3D DCNN architecture (3D Triple_UNet), slicing the experimental XCT data from multiple views (MultiView Forwarding), the i.S.Sy.Da.T.A. iterative segmentation algorithm, and (2) nonlocal means (NLM) conditioning (filtering) for the experimental XCT data. This results in good segmentation Dice scores across all phases compared to more standard approaches (i.e., standard UNet architecture, single view slicing, standard single training, and NLM conditioning). Herein, the NLM filter is replaced with the deep conditioning framework BAM SynthCOND introduced in a previous publication, which can be trained with synthetic XCT data. This leads to a significant segmentation precision increase for all phases. The proposed methods are potentially applicable to other materials and imaging techniques. KW - Automatic Segmentation KW - XCT KW - Artificial Intelligence KW - Synthetic Training Data KW - i.S.Sy.Da.T.A. KW - BAM SynthMAT KW - BAM SynthCOND KW - Triple UNet KW - Convolutional Neural Network (DCNN) PY - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-590912 SN - 1438-1656 VL - 26 IS - 2 SP - 1 EP - 9 PB - Wiley online library AN - OPUS4-59091 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Tsamos, Athanasios A1 - Evsevleev, Sergei A1 - Fioresi, R. A1 - Faglioni, F. A1 - Bruno, Giovanni T1 - Synthetic Data Generation for Automatic Segmentation of X-ray Computed Tomography Reconstructions of Complex Microstructures N2 - The greatest challenge when using deep convolutional neural networks (DCNNs) for automatic segmentation of microstructural X-ray computed tomography (XCT) data is the acquisition of sufficient and relevant data to train the working network. Traditionally, these have been attained by manually annotating a few slices for 2D DCNNs. However, complex multiphase microstructures would presumably be better segmented with 3D networks. However, manual segmentation labeling for 3D problems is prohibitive. In this work, we introduce a method for generating synthetic XCT data for a challenging six-phase Al–Si alloy composite reinforced with ceramic fibers and particles. Moreover, we propose certain data augmentations (brightness, contrast, noise, and blur), a special in-house designed deep convolutional neural network (Triple UNet), and a multi-view forwarding strategy to promote generalized learning from synthetic data and therefore achieve successful segmentations. We obtain an overall Dice score of 0.77. Lastly, we prove the detrimental effects of artifacts in the XCT data on achieving accurate segmentations when synthetic data are employed for training the DCNNs. The methods presented in this work are applicable to other materials and imaging techniques as well. Successful segmentation coupled with neural networks trained with synthetic data will accelerate scientific output. KW - Automatic segmentation KW - 3D deep convolutional neural network (3D DCNN) KW - Dice score KW - Metal matrix composite (MMC) KW - Modified U-Net architectures KW - Multi-phase materials PY - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-571243 VL - 9 IS - 2 SP - 1 EP - 23 PB - MDPI AN - OPUS4-57124 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Tsamos, Athanasios A1 - Evsevleev, Sergei A1 - Fioresi, R. A1 - Faglioni, F. A1 - Bruno, Giovanni T1 - A novel iterative algorithm to improve segmentations with deep convolutional neural networks trained with synthetic X-ray computed tomography data (i.S.Sy.Da.T.A) N2 - We propose a novel iterative segmentation algorithm (i.S.Sy.Da.T.A: Iterative Segmentation Synthetic Data Training Algorithm) employing Deep Convolutional Neural Networks and synthetic training data for X-ray tomographic reconstructions of complex microstructures. In our method, we reinforce the synthetic training data with experimental XCT datasets that were automatically segmented in the previous iteration. This strategy produces better segmentations in successive iterations. We test our algorithm with experimental XCT re constructions of a 6-phase Al-Si Matrix Composite reinforced with ceramic fibers and particles. We perform the analysis in 3D with a special network architecture that demonstrates good generalization with synthetic training data. We show that our iterative algorithm returns better segmentations compared to the standard single training approach. More specifically, phases possessing similar attenuation coefficients can be better segmented: for Al2O3 fibers, SiC particles, and Intermetallics, we see an increase of the Dice score with respect to the classic approach: from 0.49 to 0.54, from 0.66 to 0.72, and from 0.55 to 0.66 respectively. Furthermore, the overall Dice score increases from 0.77 to 0.79. The methods presented in this work are also applicable to other materials and imaging techniques. KW - Metal matrix composites (MMC) KW - Multi-phase materials KW - 3D imaging KW - Dice score KW - Automatic segmentation KW - Deep convolutional neural network (DCNN) KW - Modified U-net architectures PY - 2023 U6 - https://doi.org/10.1016/j.commatsci.2023.112112 SN - 0927-0256 VL - 223 SP - 1 EP - 8 PB - Elsevier B.V. AN - OPUS4-57482 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Tsamos, Athanasios A1 - Evsevleev, Sergei A1 - Bruno, Giovanni T1 - Noise and blur removal from corrupted X-ray computed tomography scans: A multilevel and multiscale deep convolutional framework approach with synthetic training data (BAM SynthCOND) N2 - Regardless of the experimental care practiced in acquiring X-ray computed tomography (XCT) data, artifacts might still exist, such as noise and blur. This is typical for fast XCT data acquisitions (e.g., in-situ investigations), or low-dose XCT. Such artifacts can complicate subsequent analysis of the data. Digital filters can moderately cure extensive artifacts. The selection of filter type, intensity, and order of application is not always straight forward. To tackle these problems, a complete sequential multilevel, multi-scale framework: BAM SynthCOND, employing newly designed deep convolutional neural networks (DCNNs), was formulated. Although data conditioning with neural networks is not uncommon, the main complication is that completely artifact-free XCT data for training do not exist. Thus, training data were acquired from an in-house developed library (BAM SynthMAT) capable of generating synthetic XCT material microstructures. Some novel DCNN architectures were introduced (2D/3D ACEnet_Denoise, 2D/3D ACEnet_Deblur) along with the concept of Assertive Contrast Enhancement (ACE) training, which boosts the performance of neural networks trained with continuous loss functions. The proposed methodology accomplished very good generalization from low resemblance synthetic training data. Indeed, denoising, sharpening (deblurring), and even ring artifact removal performance were achieved on experimental post-CT scans of challenging multiphase Al-Si Metal Matrix Composite (MMC) microstructures. The conditioning efficiencies were: 92% for combined denoising/sharpening, 99% for standalone denoising, and 95% for standalone sharpening. The results proved to be independent of the artifact intensity. We believe that the novel concepts and methodology developed in this work can be directly applied on the CT projections prior to reconstruction, or easily be extended to other imaging techniques such as: Microscopy, Neutron Tomography, Ultrasonics, etc. KW - XCT Data Conditioning KW - Denoising Deblurring Sharpening KW - Deep Convolutional Neural Network (DCNN) KW - Synthetic Training Data KW - ACEnet KW - Metal Matrix Composite (MMC) PY - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-579129 VL - 2 SP - 1 EP - 16 PB - Elsevier B.V. AN - OPUS4-57912 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Thiede, Tobias A1 - Mishurova, Tatiana A1 - Evsevleev, Sergei A1 - Serrano Munoz, Itziar A1 - Gollwitzer, Christian A1 - Bruno, Giovanni T1 - 3D shape analysis of powder for laser beam melting by synchrotron X-ray CT N2 - The quality of components made by laser beam melting (LBM) additive manufacturing is naturally influenced by the quality of the powder bed. A packing density <1 and porosity inside the powder particles lead to intrinsic voids in the powder bed. Since the packing density is determined by the particle size and shape distribution, the determination of these properties is of significant interest to assess the printing process. In this work, the size and shape distribution, the amount of the particle’s intrinsic porosity, as well as the packing density of micrometric powder used for LBM, have been investigated by means of synchrotron X-ray computed tomography (CT). Two different powder batches were investigated: Ti–6Al–4V produced by plasma atomization and stainless steel 316L produced by gas atomization. Plasma atomization particles were observed to be more spherical in terms of the mean anisotropy compared to particles produced by gas atomization. The two kinds of particles were comparable in size according to the equivalent diameter. The packing density was lower (i.e., the powder bed contained more voids in between particles) for the Ti–6Al–4V particles. The comparison of the tomographic results with laser diffraction, as another particle size measurement technique, proved to be in agreement. KW - Additive manufacturing KW - Laser beam melting KW - Synchrotron computed tomography KW - Powder analysis KW - Imaging PY - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-474070 SN - 2412-382X VL - 3 IS - 1 SP - 3, 1 EP - 12 PB - MDPI AN - OPUS4-47407 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Surmeneva, M. A1 - Koptyug, A. A1 - Khrapov, D. A1 - Ivanov, Yuriy A1 - Mishurova, Tatiana A1 - Evsevleev, Sergei A1 - Prymak, O. A1 - Loza, K. A1 - Epple, M. A1 - Bruno, Giovanni A1 - Surmenev, R. T1 - In situ synthesis of a binary Ti–10at% Nb alloy by electron beam melting using a mixture of elemental niobium and titanium powders N2 - This study reports the results of the preliminary assessment to fabricate Ti-10at% Nb alloy by electron beam melting (EBM®) from a blend of elemental Nb and Ti powders. The microstructure of the EBM-manufactured Ti-10at% Nb alloys is sensitive to the following factors: different sintering properties of Nb and Ti powders, powder particle properties, material viscosities at varying melt pool temperatures, β-stabilizer element content and the EBM® process parameters. Three phases were observed in as-manufactured Ti-10at% Nb alloy: μm-size Nb phase, a Nb-rich β-solid solution surrounding Nb phase, lamellar structured α-phase and β-solid solution with different distribution and volume fraction. Thus, the combination of powder particle characteristics, very short time material spends in molten condition and sluggish kinetics of mixing and diffusional process in Ti-Nb alloy results in heterogeneous microstructures depending on the local Nb content in the powder blend and the EBM® process conditions. KW - Additive manufacturing KW - Electron beam melting KW - Ti-Nb alloy KW - In situ alloying PY - 2020 U6 - https://doi.org/10.1016/j.jmatprotec.2020.116646 VL - 282 SP - 116646 PB - Elsevier B.V. AN - OPUS4-50457 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Shashev, Yury A1 - Kupsch, Andreas A1 - Lange, Axel A1 - Evsevleev, Sergei A1 - Müller, Bernd R. A1 - Osenberg, Markus A1 - Manke, Ingo A1 - Hentschel, Manfred P. A1 - Bruno, Giovanni T1 - Optimizing the visibility of X-ray phase grating interferometry N2 - The performance of grating interferometers coming up now for Imaging interfaces within materials depends on the efficiency (visibility) of their main component, namely the phase grating. Therefore, experiments with monochromatic synchrotron radiation and corresponding simulations are carried out. The visibility of a Phase grating is optimized by different photon energies, varying detector to grating distances and continuous rotation of the phase grating about the grid lines. Such kind of rotation changes the projected grating shapes, and thereby the distribution profiles of phase shifts. This yields higher visibilities than derived from ideal rectangular shapes. By continuous grating rotation and variation of the propagation distance, we achieve 2D visibility maps. Such maps provide the visibility for a certain combination of grating orientation and detector position. Optimum visibilities occur at considerably smaller distances than in the standard setup. KW - Synchrotron, BAMline KW - Talbot-Lau interferometer KW - X-ray imaging KW - X-ray refraction KW - Grating interferometry KW - X-ray phase contrast PY - 2017 U6 - https://doi.org/10.3139/120.111097 SN - 0025-5300 VL - 59 IS - 11-12 SP - 974 EP - 980 PB - Hanser Verlag AN - OPUS4-42955 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Shashev, Yury A1 - Kupsch, Andreas A1 - Lange, Axel A1 - Evsevleev, Sergei A1 - Britzke, Ralf A1 - Bruno, Giovanni A1 - Müller, Bernd R. T1 - Talbot-Lau Röntgen-Interferometrie für die zerstörungsfreie Prüfung N2 - Im vergangenen Jahrzehnt sind gitterinterferometrische Röntgenabbildungen zunehmend in den Fokus des Interesses gerückt. Sie sind insbesondere dann Vorteil, wenn der in der Standardradiographie erreichbare Kontrast nicht genügt, schwach absorbierende Mikro- und Nanostrukturen abzubilden. Auf Kosten mehrerer Aufnahmen gestattet die Talbot-Lau Gitterinterferometrie, die Beiträge aus Absorption, Refraktion und (refraktiver) Streuung zu trennen. Dazu werden die Störungen der Interferenzmuster von Phasengittern ausgewertet. Im Gegensatz zu einer Vielzahl anderer refraktionsbasierter Abbildungsmethoden kann diese Technik mit konventionellen Röntgenröhren (mit divergenter, polychromatischer Strahlung) angewendet werden. Damit ist die Technik geeignet, typische ZfP-Fragestellungen zu lösen. Hier stellen wir Untersuchungen zur Effizienz von Phasengittern vor. Die Visibilität (d.h. die Höhe der Oszillation im Interferenzmuster) wird als Funktion verschiedener geometrischer Parameter verfolgt. Eine Rotation um die Achse parallel zu den Gitterstegen ändert die Form des Gitterprofils (d.h. die Verteilung der Phasenschiebungen). Die kontinuierliche Variation des Winkels und des Detektionsabstandes führt zur Identifikation von ausgezeichneten Kombinationen mit maximaler Visibilität, wobei die Abstände deutlich geringer sind als im Standardaufbau mit senkrechter Gitterbestrahlung. Unsere Studie umfasst Simulationen für monochromatische Quelle und den Vergleich zu experimentellen Ergebnissen mit der Synchrotronstrahlung. In der Literatur wurden die Vorteile der Talbot-Lau Interferometrie für die zerstörungsfreie Prüfung von Faserkompositen und Verbundwerkstoffen demonstriert. Hier werden als Beispiele Messungen an Aluminiumtitanat (Al2TiO5) Pulver gezeigt. T2 - DGZfP 2017 CY - Koblenz, Germany DA - 22.05.2017 KW - Gitterinterferometrische Röntgenabbildung KW - Phasengitter KW - Talbot-Lau Interferometrie KW - Zerstörungsfreie Prüfung PY - 2017 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-404691 SP - 1 EP - 10 PB - DGZfP AN - OPUS4-40469 LA - deu AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Serrano Munoz, Itziar A1 - Evsevleev, Sergei A1 - Laquai, René A1 - Müller, Bernd R. A1 - Kupsch, Andreas A1 - Bruno, Giovanni T1 - Damage characterization via 2D and 3D X-ray refraction techniques N2 - We present two examples of the potential of synchrotron X-ray refraction techniques. First, we focus on the 3D imaging of hydrogen assisted cracks in an EN AW – 6060 aluminium alloy which are otherwise undetected by absorption-based CT. The second work is a quantitative analysis of the damage evolution in an Al/Al2O3 Metal Matrix Composite during interrupted in-situ tensile load. T2 - International Conference on Tomography of Materials & Structures CY - Cairns, Australia DA - 22.07.2019 KW - X-ray refraction KW - Analyzer-based imaging KW - Aluminum alloy KW - Metal matrix composite KW - Damage characterization PY - 2019 AN - OPUS4-48604 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Pauzon, C. A1 - Mishurova, Tatiana A1 - Evsevleev, Sergei A1 - Dubiez-Le Goff, S. A1 - Murugesan, S. A1 - Bruno, Giovanni A1 - Hryha, E. T1 - Residual stresses and porosity in Ti-6Al-4V produced by laser powder bed fusion as a function of process atmosphere and component design N2 - The influence of the process gas, laser scan speed, and sample thickness on the build-up of residual stresses and porosity in Ti-6Al-4V produced by laser powder bed fusion was studied. Pure argon and helium, as well as a mixture of those (30% helium), were employed to establish process atmospheres with a low residual Oxygen content of 100 ppm O2. The results highlight that the subsurface residual stresses measured by X-ray diffraction were significantly lower in the thin samples (220 MPa) than in the cuboid samples (645 MPa). This difference was attributed to the shorter laser vector length, resulting in heat accumulation and thus in-situ stress relief. The addition of helium to the process gas did not introduce additional subsurface residual stresses in the simple geometries, even for the increased scanning speed. Finally, larger deflection was found in the cantilever built under helium (after removal from the baseplate), than in those produced under argon and an argon-helium mixture. This result demonstrates that complex designs involving large scanned areas could be subjected to higher residual stress when manufactured under helium due to the gas’s high thermal conductivity, heat capacity, and thermal diffusivity. KW - Additive manufacturing KW - Ti-6Al-4V KW - Residual stress KW - Process atmosphere KW - Helium PY - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-534209 VL - 47 SP - 2340 PB - Elsevier B.V. AN - OPUS4-53420 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -