-
Predicting ptychography probe positions using single-shot phase retrieval neural network
Authors:
Ming Du,
Tao Zhou,
Jun**g Deng,
Daniel J. Ching,
Steven Henke,
Mathew J. Cherukara
Abstract:
Ptychography is a powerful imaging technique that is used in a variety of fields, including materials science, biology, and nanotechnology. However, the accuracy of the reconstructed ptychography image is highly dependent on the accuracy of the recorded probe positions which often contain errors. These errors are typically corrected jointly with phase retrieval through numerical optimization appro…
▽ More
Ptychography is a powerful imaging technique that is used in a variety of fields, including materials science, biology, and nanotechnology. However, the accuracy of the reconstructed ptychography image is highly dependent on the accuracy of the recorded probe positions which often contain errors. These errors are typically corrected jointly with phase retrieval through numerical optimization approaches. When the error accumulates along the scan path or when the error magnitude is large, these approaches may not converge with satisfactory result. We propose a fundamentally new approach for ptychography probe position prediction for data with large position errors, where a neural network is used to make single-shot phase retrieval on individual diffraction patterns, yielding the object image at each scan point. The pairwise offsets among these images are then found using a robust image registration method, and the results are combined to yield the complete scan path by constructing and solving a linear equation. We show that our method can achieve good position prediction accuracy for data with large and accumulating errors on the order of $10^2$ pixels, a magnitude that often makes optimization-based algorithms fail to converge. For ptychography instruments without sophisticated position control equipment such as interferometers, our method is of significant practical potential.
△ Less
Submitted 31 May, 2024;
originally announced May 2024.
-
AI-assisted Automated Workflow for Real-time X-ray Ptychography Data Analysis via Federated Resources
Authors:
Anakha V Babu,
Tekin Bicer,
Saugat Kandel,
Tao Zhou,
Daniel J. Ching,
Steven Henke,
SiniĊĦa Veseli,
Ryan Chard,
Antonino Miceli,
Mathew Joseph Cherukara
Abstract:
We present an end-to-end automated workflow that uses large-scale remote compute resources and an embedded GPU platform at the edge to enable AI/ML-accelerated real-time analysis of data collected for x-ray ptychography. Ptychography is a lensless method that is being used to image samples through a simultaneous numerical inversion of a large number of diffraction patterns from adjacent overlappin…
▽ More
We present an end-to-end automated workflow that uses large-scale remote compute resources and an embedded GPU platform at the edge to enable AI/ML-accelerated real-time analysis of data collected for x-ray ptychography. Ptychography is a lensless method that is being used to image samples through a simultaneous numerical inversion of a large number of diffraction patterns from adjacent overlap** scan positions. This acquisition method can enable nanoscale imaging with x-rays and electrons, but this often requires very large experimental datasets and commensurately high turnaround times, which can limit experimental capabilities such as real-time experimental steering and low-latency monitoring. In this work, we introduce a software system that can automate ptychography data analysis tasks. We accelerate the data analysis pipeline by using a modified version of PtychoNN -- an ML-based approach to solve phase retrieval problem that shows two orders of magnitude speedup compared to traditional iterative methods. Further, our system coordinates and overlaps different data analysis tasks to minimize synchronization overhead between different stages of the workflow. We evaluate our workflow system with real-world experimental workloads from the 26ID beamline at Advanced Photon Source and ThetaGPU cluster at Argonne Leadership Computing Resources.
△ Less
Submitted 9 April, 2023;
originally announced April 2023.
-
Deep learning at the edge enables real-time streaming ptychographic imaging
Authors:
Anakha V Babu,
Tao Zhou,
Saugat Kandel,
Tekin Bicer,
Zhengchun Liu,
William Judge,
Daniel J. Ching,
Yi Jiang,
Sinisa Veseli,
Steven Henke,
Ryan Chard,
Yudong Yao,
Ekaterina Sirazitdinova,
Geetika Gupta,
Martin V. Holt,
Ian T. Foster,
Antonino Miceli,
Mathew J. Cherukara
Abstract:
Coherent microscopy techniques provide an unparalleled multi-scale view of materials across scientific and technological fields, from structural materials to quantum devices, from integrated circuits to biological cells. Driven by the construction of brighter sources and high-rate detectors, coherent X-ray microscopy methods like ptychography are poised to revolutionize nanoscale materials charact…
▽ More
Coherent microscopy techniques provide an unparalleled multi-scale view of materials across scientific and technological fields, from structural materials to quantum devices, from integrated circuits to biological cells. Driven by the construction of brighter sources and high-rate detectors, coherent X-ray microscopy methods like ptychography are poised to revolutionize nanoscale materials characterization. However, associated significant increases in data and compute needs mean that conventional approaches no longer suffice for recovering sample images in real-time from high-speed coherent imaging experiments. Here, we demonstrate a workflow that leverages artificial intelligence at the edge and high-performance computing to enable real-time inversion on X-ray ptychography data streamed directly from a detector at up to 2 kHz. The proposed AI-enabled workflow eliminates the sampling constraints imposed by traditional ptychography, allowing low dose imaging using orders of magnitude less data than required by traditional methods.
△ Less
Submitted 19 September, 2022;
originally announced September 2022.
-
Thermal evolution and sintering of chondritic planetesimals III. Modelling the heat conductivity of porous chondrite material
Authors:
Stephan Henke,
Hans-Peter Gail,
Mario Trieloff
Abstract:
The construction of models for the internal constitution and the temporal evolution of large planetesimals, the parent bodies of chondrites, requires information on the heat conductivity of the complex mixture of minerals and iron metal found in chondrites. It is attempted to evaluate the heat conductivity of a multi-component mineral mixture and granular medium from the heat conductivities of its…
▽ More
The construction of models for the internal constitution and the temporal evolution of large planetesimals, the parent bodies of chondrites, requires information on the heat conductivity of the complex mixture of minerals and iron metal found in chondrites. It is attempted to evaluate the heat conductivity of a multi-component mineral mixture and granular medium from the heat conductivities of its mixture components. Random mixtures of solids with chondritic composition and packings of spheres are numerically generated. The heat conduction equation is solved in high spatial resolution for a test cube filled with such matter. From the heat flux through the cube the heat conductivity of the mixture is derived. The model results for porous material are consistent with data for compacted sandstone, but are at odds with measurements for H and L chondrites. The discrepancy is traced back to shock modification of the currently available meteoritic material by impacts on the parent body over the last 4.5 Ga. This causes numerous micro-cracks that act as additional barriers for heat transfer. The void structure in meteorites is different from that which probably existed in the pristine material of the parent bodies. The results obtained for the heat conductivity of the pristine material are used for calculating models for the evolution of the H chondrite parent body which are fitted to the cooling data of a number of H chondrites. The fit to the data good.
△ Less
Submitted 31 January, 2016;
originally announced February 2016.
-
Thermal evolution and sintering of chondritic planetesimals II. Improved treatment of the compaction process
Authors:
Hans-Peter Gail,
Stephan Henke,
Mario Trieloff
Abstract:
Reconstruction of the thermal history of individual meteorites which can be assigned to the same parent body allows to derive general characteristics of the parent body, which hold important clues on the planetary formation process. This requires to construct models for the heating of such bodies by short lived radioactives, in particular by 26Al, and its cooling by heat conduction, which then are…
▽ More
Reconstruction of the thermal history of individual meteorites which can be assigned to the same parent body allows to derive general characteristics of the parent body, which hold important clues on the planetary formation process. This requires to construct models for the heating of such bodies by short lived radioactives, in particular by 26Al, and its cooling by heat conduction, which then are compared with the cooling histories of the meteorites. The heat conductivity of the planetesimal material depends critically on the porosity of the chondritic material and changes by sintering of the material at elevated temperatures and pressures. Compaction of a granular material is a key process for the thermal history of the parent bodies of meteorites. The modelling of the compaction process is improved by applying concepts originally developed for hot isostatic pressing in metallurgical processes. It is extended to a binary mixture of matrix and chondrules, as observed in chondrites. By comparison with published data on sintering experiments it is shown that the algorithm allows a sufficiently accurate modelling of the compaction of silicate material. It is shown that the characteristic temperature at which sintering occurs is different for matrix or chondrule dominated precursor material. We apply the new method to model the thermal evolution of the parent body of the H chondrites and determine an improved optimized set of model parameters for this body.
△ Less
Submitted 11 November, 2014;
originally announced November 2014.
-
Perfect and Stable Hybrid Glasses from Strong and Fragile Metal-Organic Framework Liquids
Authors:
T. D. Bennett,
**-Chong Tan,
Y. Z. Yue,
C. Ducati,
N. Terril,
H. H. M. Yeung,
Z. Zhou,
W. Chen,
S. Henke,
A. K. Cheetham,
G. N. Greaves
Abstract:
Hybrid glasses connect emerging fields of metal-organic frameworks (MOFs) with the glass-formation, amorphization, and melting processes of these structurally diverse and chemically versatile systems. Most zeolites, including MOFs, amorphize around the glass transition, devitrifying and then melting at much higher temperatures. The relationship between the two processes has so far not been investi…
▽ More
Hybrid glasses connect emerging fields of metal-organic frameworks (MOFs) with the glass-formation, amorphization, and melting processes of these structurally diverse and chemically versatile systems. Most zeolites, including MOFs, amorphize around the glass transition, devitrifying and then melting at much higher temperatures. The relationship between the two processes has so far not been investigated. Herein we show how heating first results in a low density perfect glass, following an order-order transition, leading to a super-strong liquid of low fragility that dynamically controls MOF collapse. A subsequent order-disorder transition creates a high density liquid of greater fragility. After crystallization and melting, subsequent cooling results in a stable glass virtually identical to the high density phase. Furthermore, the wide-ranging melting temperatures of different MOFs suggest these can be differentiated by topology. Our research provides new insight into the stability and functionality of these novel ductile crystalline materials, including the possibility of melt-casting MOFs.
△ Less
Submitted 13 September, 2014;
originally announced September 2014.
-
Thermal history modeling of the H chondrite parent body
Authors:
Stephan Henke,
Hans-Peter Gail,
Mario Trieloff,
Winfried H. Schwarz,
Thorsten Kleine
Abstract:
The cooling histories of individual meteorites can be empirically reconstructed by using ages from different radioisotopic chronometers with distinct closure temperatures. For a group of meteorites derived from a single parent body such data permit the reconstruction of the cooling history and properties of that body. Particularly suited are H chondrites because precise radiometric ages over a wid…
▽ More
The cooling histories of individual meteorites can be empirically reconstructed by using ages from different radioisotopic chronometers with distinct closure temperatures. For a group of meteorites derived from a single parent body such data permit the reconstruction of the cooling history and properties of that body. Particularly suited are H chondrites because precise radiometric ages over a wide range of closure temperatures are available. A thermal evolution model for the H chondrite parent body is constructed by using all H chondrites for which at least three different radiometric ages are available. Several key parameters determining the thermal evolution of the H chondrite parent body and the unknown burial depths of the H chondrites are varied until an optimal fit is obtained. The fit is performed by an 'evolution algorithm'. Empirical data for eight samples are used for which radiometric ages are available for at least three different closure temperatures. A set of parameters for the H chondrite parent body is found that yields excellent agreement (within error bounds) between the thermal evolution model and empirical data of six of the examined eight chondrites. The new thermal model constrains the radius and formation time of the H chondrite parent body (possibly (6) Hebe), the initial burial depths of the individual H chondrites, the average surface temperature of the body, the average initial porosity of the material the body accreted from, and the initial 60Fe content of the H chondrite parent body.
△ Less
Submitted 22 August, 2012;
originally announced August 2012.
-
Thermal evolution and sintering of chondritic planetesimals
Authors:
Stephan Henke,
Hans-Peter Gail,
Mario Trieloff,
Winfried H. Schwarz,
Thorsten Kleine
Abstract:
The major aim of this study is to assess the effects of sintering of initially porous material on the thermal evolution of planetesimals, and to constrain the values of basic parameters that determined the structure and evolution of the H chondrite parent body. A new code is presented for modeling the thermal evolution of ordinary chondrite parent bodies that initially are highly porous and underg…
▽ More
The major aim of this study is to assess the effects of sintering of initially porous material on the thermal evolution of planetesimals, and to constrain the values of basic parameters that determined the structure and evolution of the H chondrite parent body. A new code is presented for modeling the thermal evolution of ordinary chondrite parent bodies that initially are highly porous and undergo sintering by hot pressing as they are heated by decay of radioactive nuclei. The pressure and temperature stratification in the interior of the bodies is calculated by solving the equations of hydrostatic equilibrium and energy transport. The decrease of porosity of the granular material by hot pressing due to self-gravity is followed by solving a set of equations for the sintering of powder materials. For the heat conductivity of granular material we combine recently measured data for highly porous powder materials, relevant for the surface layers of planetesimals, with data for heat conductivity of chondrite material, relevant for the strongly sintered material in deeper layers. To demonstrate the capability of our new model, the thermal evolution of the H chondrite parent body was reconstructed. The model starts with a porous body that is later compacted first by 'cold pressing' at low temperatures and then by 'hot pressing' for temperatures above \approx 700 K, i.e., the threshold temperature for sintering of silicates. The thermal model was fitted to the well constrained cooling histories of the two H chondrites Kernouve (H6) and Richardton (H5).
△ Less
Submitted 21 October, 2011;
originally announced October 2011.
-
Inertial particle entrainment in a shearless mixing layer
Authors:
Peter J. Ireland,
Martin S. Henke,
Lance R. Collins
Abstract:
This article accompanies the submission of a fluid dynamics video (entry V032) of inertial particle entrainment in a shearless mixing layer for the Gallery of Fluid Motion of the 64th Annual Meeting of the American Physical Society, Division of Fluid Dynamics.
This article accompanies the submission of a fluid dynamics video (entry V032) of inertial particle entrainment in a shearless mixing layer for the Gallery of Fluid Motion of the 64th Annual Meeting of the American Physical Society, Division of Fluid Dynamics.
△ Less
Submitted 13 October, 2011;
originally announced October 2011.
