Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “algorithm complexity analysis”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 883 records · Page 49
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

GMOSS: ALL-SKY MODEL OF SPECTRAL RADIO BRIGHTNESS BASED ON PHYSICAL COMPONENTS AND ASSOCIATED RADIATIVE PROCESSES

We present the Global Model for the Radio Sky Spectrum (GMOSS), a novel, physically motivated model of the low-frequency radio sky from 22 MHz to 23 GHz. GMOSS invokes different physical components and associated radiative processes to describe the sky spectrum over 3072 pixels of 5° resolution. The spectra are allowed to be convex, concave, or of more complex form with contributions from synchrotron emission, thermal emission, and free–free absorption included. Physical parameters that describe the model are optimized to best fit four all-sky maps at 150 MHz, 408 MHz, 1420 MHz, and 23 GHz and two maps at 22 and 45 MHz generated using the Global Sky Model of de Oliveira-Costa et al. The fractional deviation of the model from data has a median value of 6% and is less than 17% for 99% of the pixels. Though aimed at the modeling of foregrounds for the global signal arising from the redshifted 21 cm line of hydrogen during the Cosmic Dawn and the Epoch of Reionization (EoR), over redshifts 150≲z≲6, GMOSS is well suited for any application that requires simulating spectra of the low-frequency radio sky as would be observed by the beam of any instrument. The complexity in spectral structure that naturally arises from the underlying physics of the model provides a useful expectation for departures from smoothness in EoR foreground spectra and hence may guide the development of algorithms for EoR signal detection. This aspect is further explored in a subsequent paper.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1991

Hidden Markov model analysis of force/torque information in telemanipulation

A model for the prediction and analysis of sensor information recorded during robotic performance of telemanipulation tasks is presented. The model uses the hidden Markov model to describe the task structure, the operator's or intelligent controller's goal structure, and the sensor signals. A methodology for constructing the model parameters based on engineering knowledge of the task is described. It is concluded that the model and its optimal state estimation algorithm, the Viterbi algorithm, are very succesful at the task of segmenting the data record into phases corresponding to subgoals of the task. The model provides a rich modeling structure within a statistical framework, which enables it to represent complex systems and be robust to real-world sensory signals.

Hannaford, Blake↗
NASA NTRS2014

Snowfall Rate Retrieval using NPP ATMS Passive Microwave Measurements

Passive microwave measurements at certain high frequencies are sensitive to the scattering effect of snow particles and can be utilized to retrieve snowfall properties. Some of the microwave sensors with snowfall sensitive channels are Advanced Microwave Sounding Unit (AMSU), Microwave Humidity Sounder (MHS) and Advance Technology Microwave Sounder (ATMS). ATMS is the follow‐on sensor to AMSU and MHS. Currently, an AMSU and MHS based land snowfall rate (SFR) product is running operationally at NOAA/NESDIS. Based on the AMSU/MHS SFR, an ATMS SFR algorithm has been developed recently. The algorithm performs retrieval in three steps: snowfall detection, retrieval of cloud properties, and estimation of snow particle terminal velocity and snowfall rate. The snowfall detection component utilizes principal component analysis and a logistic regression model. The model employs a combination of temperature and water vapor sounding channels to detect the scattering signal from falling snow and derive the probability of snowfall (Kongoli et al., 2014). In addition, a set of NWP model based filters is also employed to improve the accuracy of snowfall detection. Cloud properties are retrieved using an inversion method with an iteration algorithm and a two‐stream radiative transfer model (Yan et al., 2008). A method developed by Heymsfield and Westbrook (2010) is adopted to calculate snow particle terminal velocity. Finally, snowfall rate is computed by numerically solving a complex integral. The ATMS SFR product is validated against radar and gauge snowfall data and shows that the ATMS algorithm outperforms the AMSU/MHS SFR.

Meng, Huan↗
DOE OSTI2018

Computational complexity of the landscape II—Cosmological considerations

We propose a new approach for multiverse analysis based on computational complexity, which leads to a new family of “computational” measure factors. By defining a cosmology as a space–time containing a vacuum with specified properties (for example small cosmological constant) together with rules for how time evolution will produce the vacuum, we can associate global time in a multiverse with clock time on a supercomputer which simulates it. We argue for a principle of “limited computational complexity” governing early universe dynamics as simulated by this supercomputer, which translates to a global measure for regulating the infinities of eternal inflation. The rules for time evolution can be thought of as a search algorithm, whose details should be constrained by a stronger principle of “minimal computational complexity”. Unlike previously studied global measures, ours avoids standard equilibrium considerations and the well-known problems of Boltzmann Brains and the youngness paradox. We also give various definitions of the computational complexity of a cosmology, and argue that there are only a few natural complexity classes.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

RE-INTEGRATE EMT Simulation Software: Graph Convolutional Network for Sparse Matrix Pattern Detection

The increasing complexity of power networks, driven by proliferation of inverters, presents analytical challenges that simplified models often fail to capture, necessitating Electromagnetic Transient (EMT) simulations. EMT models are represented as discretized differential-algebraic equations (DAEs), forming a linear system Ax = b that is computationally intensive to solve. Due to inherent sparsity of adjacency matrix A, distinct patterns emerge that, when accurately identified, enable efficient solver selection to minimize computation time. However, identifying ideal pattern is complicated by numerous reordering algorithms and limited structural insights. To address this, we introduce a Graph Convolutional Network (GCN) model for classifying sparse matrix patterns common in power system analysis. The model, achieving 96% test accuracy, is validated using PV plant models of 125 MW capacities connected to New England 39-bus transmission system (TS), and further scaled to a 4,992-bus network with 384 PV plants, yielding 191, 616 × 191, 616 sized A matrix. For all cases, the GCN model accurately identifies the matrix’s intrinsic sparse pattern, demonstrating its potential to enhance solver performance in EMT analysis.

Hossain, Md Rifat [Florida International Universit↗
DOE OSTI2026

Scalable Computation of Topological Abstractions for Scalar Data

Topological data analysis has become an important tool for large scale scalar data analysis and visualization, efficiently extracting the inherent structure and features of interest of the data. However, with growing dataset sizes and complexity, it is increasingly becoming infeasible to compute topological abstractions of interest in serial and on single machines. This paper presents the state of the art in the scalable computation of topological abstractions on scalar data, in shared memory parallel on single machines, and in distributed memory parallel on multiple machines. We highlight results for set‐based, graph‐based and complex‐based abstractions and organize the state of the art based on this taxonomy. The paper identifies parallelization and distribution techniques common in topological algorithms and highlights further areas of interest with underdeveloped efforts.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2018

Decadal Seasonal Shifts of Precipitation and Temperature in TRMM and AIRS Data

We present results from an analysis of seasonal phase shifts in the global precipitation and surface temperatures. We use data from the TRMM (Tropical Rainfall Measuring Mission) Multi-satellite Precipitation Algorithm (TMPA), and the Atmospheric Infrared Sounder (AIRS) on Aqua satellite, all hosted at NASA Goddard Earth Science Data and Information Services Center (GES DISC). We explore the information content and data usability by first aggregating daily grids from the entire records of both missions to pentad (5-day) series which are then processed using Singular Value Decomposition approach. A strength of this approach is the normalized principal components that can then be easily converted from real to complex time series. Thus, we can separate the most informative, the seasonal, components and analyze unambiguously for potential seasonal phase drifts. TMPA and AIRS records represent correspondingly 20 and 15 years of data, which allows us to run simple “phase learning†from the first 5 years of records and use it as reference. The most recent 5 years are then phase-compared with the reference. We demonstrate that the seasonal phase of global precipitation and surface temperatures has been stable in the past two decades. However, a small global trend of delayed precipitation, and earlier arrival of surface temperatures seasons, are detectable at 95% confidence level. Larger phase shifts are detectable at regional level, in regions recognizable from the Eigen vectors to having strong seasonal patterns. For instance, in Central North America, including the North American Monsoon region, confident phase shifts of 1-2 days per decade are detected at 95% confidence level. While seemingly symbolic, these shifts are indicative of larger changes in the Earth Climate System. We thus also demonstrate a potential usability scenario of Earth Science Data Records curated at the NASA GES DISC in partnership with Earth Science Missions.

surface temperatures↗
NASA NTRS2022

The Harmonic Linearized Navier-Stokes Equations for Transition Prediction in Three-Dimensional Flows

The conventional method to predict the onset of laminar-turbulent transition in convectively unstable boundary-layer flows is based on the logarithmic amplification ratio, the so-called N-factor, of the linear instability waves. To calculate the N-factor, the flow variables are decomposed into a laminar basic state solution and the linear disturbances, which are assumed to be harmonic in time. The most commonly used linear stability analysis approaches include the locally parallel linear stability theory (LST) and the non-local, weakly nonparallel parabolized stability equations (PSE). However, these methods do not account for strong streamwise gradients that are encountered in several configurations of interest, as roughness elements, steps, gaps, or corners. To solve the linear evolution of disturbances along such strongly nonparallel regions, the harmonic linearized Navier-Stokes equations (HLNSE) need to be solved. The discretization of the HLNSE for spanwise/azimuthally inhomogeneous laminar basic states yields a linear system of complex arithmetic with a leading dimension of the order of 107 to 108. A combined multithread and multiprocessor algorithm is implemented for the direct solution of such linear system. Results for a supersonic boundary layer over a three-dimensional roughness patch show good agreement with experimental measurements when the evolution of the instability waves over the roughness patch is included via the HLNSE.

Boundary Layer Stability↗
DOE OSTI2016

Dynamic Multi-Sensor Multi-Mission Optimal Planning Tool

Remote sensing systems have firmly established a role in providing immense value to commercial industry, scientific exploration, and national security. Continued maturation of sensing technology has reduced the cost of deploying highly-capable sensors while at the same time increased reliance on the information these sensors can provide. The demand for time on these sensors is unlikely to diminish. Coordination of next-generation sensor systems, larger constellations of satellites, unmanned aerial vehicles, ground telescopes, etc. is prohibitively complex for existing heuristics-based scheduling techniques. The project was a two-year collaboration spanning multiple Sandia centers and included a partnership with Texas A&M University. We have developed algorithms and software for collection scheduling, remote sensor field-of-view pointing models, and bandwidth-constrained prioritization of sensor data. Our approach followed best practices from the operations research and computational geometry communities. These models provide several advantages over state of the art techniques. In particular, our approach is more flexible compared to heuristics that tightly couple models and solution techniques. First, our mixed-integer linear models afford a rigorous analysis so that sensor planners can quantitatively describe a schedule relative to the best possible. Optimal or near-optimal schedules can be produced with commercial solvers in operational run-times. These models can be modified and extended to incorporate different scheduling and resource constraints and objective function definitions. Further, we have extended these models to proactively schedule sensors under weather and ad hoc collection uncertainty. This approach stands in contrast to existing deterministic schedulers which assume a single future weather or ad hoc collection scenario. The field-of-view pointing algorithm produces a mosaic with the fewest number of images required to fully cover a region of interest. The bandwidth-constrained algorithms find the highest priority information that can be transmitted. All of these are based on mixed-integer linear programs so that, in the future, collection scheduling, field-of-view, and bandwidth prioritization can be combined into a single problem. Experiments conducted using the developed models, commercial solvers, and benchmark datasets have demonstrated that proactively scheduling against uncertainty regularly and significantly outperforms deterministic schedulers.

47 OTHER INSTRUMENTATION↗
NASA NTRS1992

A High Angle of Attack Inviscid Shuttle Orbiter Computation

An accurate description of the aerothermal environment is required to minimize the weight of the Thermal Protection System required on the leeside of winged reentry vehicles. The inability of ground-test facilities to reproduce the high enthalpy, separated flow present during reentry flight conditions, coupled with the prohibitive expense of flight tests, leads to the use of an analytical method - namely Computational Fluid Dynamics (CFD) - to describe the flow. While the ultimate goal of this work is to accurately predict the leeside flow and its associated thermal environment, an essential and reasonable first step towards that goal is a comparison of pressure predictions by a code with wind-tunnel data. Until such CFD pressure predictions agree with wind-tunnel test cases, there is little hope of accurately predicting the thermal environment at flight conditions. Thus, the objective of this study is to compare the pressure distributions predicted by inviscid, perfect gas CFD to Shuttle Orbiter wind-tunnel data and to address any significant issues encountered during the computation. While flight data is available for the Shuttle Orbiter, a wind-tunnel case is chosen for this study to allow a tractable problem for preliminary investigation. A wind-tunnel case allows the perfect gas assumption for the flow chemistry. This provides a significant computational savings over a several species finite-rate chemistry model which would be necessary if high-temperature effects present at flight conditions were to be included. In addition, by concentrating on the surface pressures, the analysis need only consider inviscid flow for general evaluation of the code capability. This further reduces the computational expense due to the absence of viscous terms and the associated decrease in the number of points required for the computational grid. Previous computational efforts (such as STEIN and HALIS) have been directed toward the windward surface quantities, primarily due to restrictions in treating either the winged geometry or its associated subsonic regions at high angle of attack. The code used for this study, the LAURA (Langley Aerothermodynamic Upwind Relaxation Algorithm) code of Gnoffo, represents a state-of-the-art code for computing the flow over complex configurations at hypersonic speeds. In the study, the LAURA code is applied to a wind-tunnel condition to initiate the assessment of the code's ability to predict the flow over a relatively complex hypersonic vehicle at high angles of attack. This presentation is used to highlight the pertinent results of a more detailed investigation of the inviscid calculation over the Shuttle Orbiter with the LAURA code.

William L. Kleb↗
NASA NTRS2013

Validating an Air Traffic Management Concept of Operation Using Statistical Modeling

Validating a concept of operation for a complex, safety-critical system (like the National Airspace System) is challenging because of the high dimensionality of the controllable parameters and the infinite number of states of the system. In this paper, we use statistical modeling techniques to explore the behavior of a conflict detection and resolution algorithm designed for the terminal airspace. These techniques predict the robustness of the system simulation to both nominal and off-nominal behaviors within the overall airspace. They also can be used to evaluate the output of the simulation against recorded airspace data. Additionally, the techniques carry with them a mathematical value of the worth of each prediction-a statistical uncertainty for any robustness estimate. Uncertainty Quantification (UQ) is the process of quantitative characterization and ultimately a reduction of uncertainties in complex systems. UQ is important for understanding the influence of uncertainties on the behavior of a system and therefore is valuable for design, analysis, and verification and validation. In this paper, we apply advanced statistical modeling methodologies and techniques on an advanced air traffic management system, namely the Terminal Tactical Separation Assured Flight Environment (T-TSAFE). We show initial results for a parameter analysis and safety boundary (envelope) detection in the high-dimensional parameter space. For our boundary analysis, we developed a new sequential approach based upon the design of computer experiments, allowing us to incorporate knowledge from domain experts into our modeling and to determine the most likely boundary shapes and its parameters. We carried out the analysis on system parameters and describe an initial approach that will allow us to include time-series inputs, such as the radar track data, into the analysis

Statistical emulation↗
NASA NTRS2008

Cassini Information Management System in Distributed Operations Collaboration and Cassini Science Planning

Launched on October 15, 1997, the Cassini-Huygens spacecraft began its ambitious journey to the Saturnian system with a complex suite of 12 scientific instruments, and another 6 instruments aboard the European Space Agencies Huygens Probe. Over the next 6 1/2 years, Cassini would continue its relatively simplistic cruise phase operations, flying past Venus, Earth, and Jupiter. However, following Saturn Orbit Insertion (SOI), Cassini would become involved in a complex series of tasks that required detailed resource management, distributed operations collaboration, and a data base for capturing science objectives. Collectively, these needs were met through a web-based software tool designed to help with the Cassini uplink process and ultimately used to generate more robust sequences for spacecraft operations. In 2001, in conjunction with the Southwest Research Institute (SwRI) and later Venustar Software and Engineering Inc., the Cassini Information Management System (CIMS) was released which enabled the Cassini spacecraft and science planning teams to perform complex information management and team collaboration between scientists and engineers in 17 countries. Originally tailored to help manage the science planning uplink process, CIMS has been actively evolving since its inception to meet the changing and growing needs of the Cassini uplink team and effectively reduce mission risk through a series of resource management validation algorithms. These algorithms have been implemented in the web-based software tool to identify potential sequence conflicts early in the science planning process. CIMS mitigates these sequence conflicts through identification of timing incongruities, pointing inconsistencies, flight rule violations, data volume issues, and by assisting in Deep Space Network (DSN) coverage analysis. In preparation for extended mission operations, CIMS has also evolved further to assist in the planning and coordination of the dual playback redundancy of highvalue data from targets such as Titan and Enceladus. This paper will outline the critical role that CIMS has played for Cassini in the distributed ops paradigm throughout operations. This paper will also examine the evolution that CIMS has undergone in the face of new science discoveries and fluctuating operational needs. And finally, this paper will conclude with theoretical adaptation of CIMS for other projects and the potential savings in cost and risk reduction that could potentially be tapped into by future missions.

Equils, Douglas J.↗
DOE OSTI2025

Tutorial: Machine-Learning-Based CREASE-2D Analysis of 2D SAXS Profiles to Characterize Anisotropic Nanostructures in Soft Materials

We present a tutorial to guide users on how to extend the Computational Reverse Engineering Analysis of Scattering Experiments-2D (CREASE-2D) framework to interpret their experimental two-dimensional small-angle scattering (SAS) data from soft materials (e.g., polymers, peptide amphiphiles, biomolecular fibrils). Unlike most traditional SAS analysis approaches, which typically rely on azimuthally averaged onedimensional (1D) profiles, CREASE-2D utilizes the complete 2D scattering profile to reveal information about anisotropy in the structure. In past applications, CREASE has provided insights into complex structural features, including the cross-sectional shapes of assembled nanostructures and dispersity in these features, which are difficult to discern with existing analytical models. While (1D- ) CREASE has been applied to SANS and SAXS data, this tutorial shares the steps for implementing CREASE-2D using an example of a dipeptide solution system, for which we have SAXS data. We present details for these steps involved in using CREASE-2D to interpret SAXS profiles: how to preprocess SAXS data, define relevant structural features, generate three-dimensional real-space structures for specific values of these features, train a machine learning (ML) surrogate model to predict scattering profiles for given structural features, and optimize these features using genetic algorithms (GA). Then, we use these steps to interpret complex 2DSAXS data collected from dipeptide solutions that, in microscopy images, exhibit nanoscale structures that could be elliptical tubes/ flat tapes/cylinders or a combination of these cross sections. Open-source codes, computational hardware, and software requirements, as well as the strengths and limitations of this protocol, are also presented. We expect researchers working with (soft) biomaterials, peptide amphiphiles, amphiphilic polymer solutions, polymer nanocomposites, and blends of particles/polymers will find this CREASE-2D method and this tutorial of use.

CREASE↗
DOE OSTI2024

Quantum-parallel vectorized data encodings and computations on trapped-ion and transmon QPUs

Compact data representations in quantum systems are crucial for the development of quantum algorithms for data analysis. In this study, we present two innovative data encoding techniques, known as QCrank and QBArt, which exhibit significant quantum parallelism via uniformly controlled rotation gates. The QCrank method encodes a series of real-valued data as rotations on data qubits, resulting in increased storage capacity. On the other hand, QBArt directly incorporates a binary representation of the data within the computational basis, requiring fewer quantum measurements and enabling well-established arithmetic operations on binary data. We showcase various applications of the proposed encoding methods for various data types. Notably, we demonstrate quantum algorithms for tasks such as DNA pattern matching, Hamming weight computation, complex value conjugation, and the retrieval of a binary image with 384 pixels, all executed on the Quantinuum trapped-ion QPU. Furthermore, we employ several cloud-accessible QPUs, including those from IBMQ and IonQ, to conduct supplementary benchmarking experiments.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Automatic Boundary-Layer Adaptation of Structured Grids in VULCAN-CFD

In supersonic and hypersonic flow computations, well-resolved boundary layers are essential for accurate quantification of surface heating and transition prediction, particularly via linear stability analysis. Grid design for hypersonic flows with shocks, boundary-layer separation, and/or complex mean flow features incorporating spanwise/azimuthal inhomogeneities is a difficult issue. In comparison to a manual grid adaptation procedure, an autonomous grid adaptation technique offers significant improvements in computing time and solution quality. The VULCAN-CFD solver already includes a validated procedure for automatic adaption of structured grids to the bow shock. The present focus is on implementing an automatic boundary-layer adaptation capability in VULCAN-CFD that adapts structured, multiblock grids to both the bow shock and the boundary layer at the same time. The boundary-layer adaptation algorithm allows the user to specify the number of cells within the boundary layer, along with the input parameters used for detecting the edge of the boundary layer, namely, the variable used in the edge detection criterion,the edge detection method, the detection direction, and the relaxation factor used during the morphing of the grid. The algorithm automatically distributes grid points along the wall-normal direction to achieve a smooth variation in grid spacing from the edge of the boundary layer to a"junction" location within the outer part of the grid. Illustrative results are presented for three different high-speed configurations: the two-dimensional flow over a cylinder at Mach 17.6 and unit Reynolds number of Re=0.38x10^6 m^-1, the axisymmetric flow over a cone-cylinder-flare model at Mach 6.0 and Re = 10.5×10^6 m^-1, and the three-dimensional flow over a blunt, 7-degree half-angle cone at 5-degree angle of attack in a Mach 9.79 flow with Re = 17.1×10^6 m^-1. The automated boundary-layer adaptation is shown to provide an adequate grid topology that is aligned with the bow shock in the outer part of the grid and also resolves the viscous boundary-layer region close to the surface.

boundary layer transition↗
DOE OSTI2017

Analysis of the VERA Core Physics Benchmark Problems using cosRMC Code

In reactor physics, two kinds of numerical methods are generally employed to solve the neutron transport equations, one of which is the deterministic method and the other one is the Monte Carlo method. Monte Carlo is widely used to provide reference solution because of its ability to deal with continuous energy cross sections and complex geometry. Just like any statistical process, the Monte Carlo method requires repetition to achieve a small relative uncertainty, and therefore may inevitably have larger simulation times than deterministic method. Fortunately, the great advancements of computer technology make it possible to apply Monte Carlo method in reactor core analysis. To promote nuclear power plant system design and simulation software self-reliance development, a Monte Carlo neutron/photon transport code cosRMC has been developed jointly by State Nuclear Power Software Development Center (SNPSDC) and Tsinghua University. The cosRMC is aimed at achieving full core calculations and analysis with high fidelity and efficiency by means of advanced methodologies and algorithms as well as high performance computing techniques. The cosRMC has many functions as criticality calculation, fixed-source calculation, burnup calculation, etc. In order to ensure its reliability, several types of verification and validation (V and V) work have been and are being carried out for cosRMC. The simulation results of cosRMC based on Problem 1 to 3 of VERA Core Physics Benchmark showed its good capability of 2D lattice and 3D assembly modeling. In this paper, some progressive verification results of multi-assembly and full core problems are presented.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Deep Learning Segmentation of Complex Features in Atomic-Resolution Phase-Contrast Transmission Electron Microscopy Images

Phase-contrast transmission electron microscopy (TEM) is a powerful tool for imaging the local atomic structure of materials. TEM has been used heavily in studies of defect structures of two-dimensional materials such as monolayer graphene due to its high dose efficiency. However, phase-contrast imaging can produce complex nonlinear contrast, even for weakly scattering samples. It is, therefore, difficult to develop fully automated analysis routines for phase-contrast TEM studies using conventional image processing tools. For automated analysis of large sample regions of graphene, one of the key problems is segmentation between the structure of interest and unwanted structures such as surface contaminant layers. In this study, we compare the performance of a conventional Bragg filtering method with a deep learning routine based on the U-Net architecture. We show that the deep learning method is more general, simpler to apply in practice, and produces more accurate and robust results than the conventional algorithm. We provide easily adaptable source code for all results in this paper and discuss potential applications for deep learning in fully automated TEM image analysis.

47 OTHER INSTRUMENTATION↗