Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “image recognition”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Defect Recognition for Eddy Current Testing of Spent Nuclear Fuel Canister using Convolutional Neural Network

This paper proposes an accurate and robust defect detection solution for 304L and 306L stainless steel (SS) weld. In the proposed solution, Eddy current testing (ECT) is employed to generate 2-dimensional (2D) data for samples under test with defects. The 2D data can be treated as images for deep learning-based defect detection. Since convolutional neural networks (CNNs) are powerful in processing images, CNN is employed in this study for defect detection. Experiments are conducted on a submerged arc welding (SAW) 304L SS weld sample with an artificial crack generated by waterjet cutting. The ECT data on this seeded fault sample is utilized to verify the proposed solution. For this purpose, the ECT measurement are separated as from Fault area and Normal area, which are used for CNN training. After training, the testing data is used for verification. Experimental results demonstrate the feasibility and effectiveness of the proposed solution.

Niu, Guangxing↗
DOE OSTI2026

Improving ICARUS track reconstruction algorithms

The ICARUS experiment is part of the Short-Baseline Neutrino program at Fermilab. Its primary objective is to explore the possible existence of sterile neutrinos in the O(1 eV) mass range and to clarify the anomalies observed in the Liquid Scintillator Neutrino Detector and MiniBooNE experiments. The ICARUS-T600 detector is a Liquid Argon Time Projection Chamber, capable of producing high-resolution 3D images and precise calorimetric measurements of ionizing particles. This technology allows for a detailed study of neutrino interactions across a broad energy range, from a few keV to several hundred GeV. The track reconstruction is achieved through a software framework that applies a series of pattern recognition algorithms, transforming raw detector signals into fully reconstructed event topologies. This process involves identifying interaction vertices, particle tracks, and electromagnetic showers within the TPC. However, in certain cases, these algorithms may mistakenly break a single particle track into several shorter segments, interpreting each as a distinct particle. Since track length is used to estimate the particle's energy, such fragmentation can result in an energy underestimation of several hundred MeV. Furthermore, when a track is split into multiple segments, the particle identification (which relies on analyzing the energy loss as a function of the residual range) may fail, potentially leading to the loss of the entire event. To mitigate this problem, we have developed a dedicated algorithm designed to identify and reconnect (“stitch”) the tracks that were erroneously divided into multiple segments.

Ricci, Alessandro Maria [Pisa U.; INFN, Pisa] (ORC↗
DOE OSTI2023

Recognizability of Demographically Altered Computerized Facial Approximations in an Automated Facial Recognition Context for Potential Application in Unidentified Persons Data Repositories

This study examined the recognizability of demographically altered facial approximations for potential utility in unidentified persons tracking systems. Five computer-generated approximations were generated for each of 26 African male participants using the following demographic parameters: (i) African male (true demographics), (ii) African female, (iii) Caucasian male, (iv) Asian male, and (v) Hispanic male. Overall, 62% of the true demographic facial approximations for the 26 African male participants examined were matched to a corresponding life photo within the top 50 images of a candidate list generated from an automated blind search of an optimally standardized gallery of 6159 photographs. When the African male participants were processed as African females, the identification rate was 50%. In contrast, less congruent identification rates were observed when the African male participants were processed as Caucasian (42%), Asian (35%), and Hispanic (27%) males. The observed results suggest that approximations generated using the opposite sex may be operationally informative if sex is unknown. The performance of approximations generated using alternative ancestry assignments, however, was less congruent with the performance of the true demographic approximation (African male) and may not yield as operationally constructive data as sex-altered approximations.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Figure Descriptive Text Extraction Using Ontological Representation

Abstract Experimental research publications provide figure form resources including graphs, charts, and any type of images to effectively support and convey methods and results. To describe figures, authors add captions, which are often incomplete, and more descriptions reside in body text. This work presents a method to extract figure descriptive text from the body of scientific articles. We adopted ontological semantics to aid concept recognition of figure-related information, which generates human- and machine-readable knowledge representations from sentences. Our results show that conceptual models bring an improvement in figure descriptive sentence classification over word-based approaches.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Experimental Observations of the Topology of Convolutional Neural Network Activations

Topological data analysis (TDA) is a branch of computational mathematics, bridging algebraic topology and data science, that provides compact, noise-robust representations of complex structures. Deep neural networks (DNNs) learn millions of parameters associated with a series of transformations defined by the model architecture resulting in high-dimensional, difficult to interpret internal representations of input data. As DNNs become more ubiquitous across multiple sectors of our society, there is increasing recognition that mathematical methods are needed to aid analysts, researchers, and practitioners in understanding and interpreting how these models' internal representations relate to the final classification. In this paper we apply cutting edge techniques from TDA with the goal of gaining insight towards interpretability of convolutional neural networks used for image classification. We use two common TDA approaches to explore several methods for modeling hidden layer activations as high-dimensional point clouds, and provide experimental evidence that these point clouds capture valuable structural information about the model's process. First, we demonstrate that a distance metric based on persistent homology can be used to quantify meaningful differences between layers and discuss these distances in the broader context of existing representational similarity metrics for neural network interpretability. Second, we show that a mapper graph can provide semantic insight as to how these models organize hierarchical class knowledge at each layer. These observations demonstrate that TDA is a useful tool to help deep learning practitioners unlock the hidden structures of their models.

topological data analysis, deep learning↗
DOE OSTI2023

Status of DUNE Offline Computing

We summarize the status of Deep Underground Neutrino Experiment (DUNE) Offline Software and Computing program. We describe plans for the computing infrastructure needed to acquire, catalog, reconstruct, simulate and analyze the data from the DUNE experiment and its prototypes in pursuit of the experiment's physics goals of precision measurements of neutrino oscillation parameters, detection of astrophysical neutrinos, measurement of neutrino interaction properties and searches for physics beyond the Standard Model. In contrast to traditional HEP computational problems, DUNE's Liquid Argon Time Projection Chamber data consist of simple but very large (many GB) data objects which share many characteristics with astrophysical images. We have successfully reconstructed and simulated data from 4% prototype detector runs at CERN. The data volume from the full DUNE detector, when it starts commissioning late in this decade will present memory management challenges in conventional processing but significant opportunities to use advances in machine learning and pattern recognition as a frontier user of High Performance Computing facilities capable of massively parallel processing. Our goal is to develop infrastructure resources that are flexible and accessible enough to support creative software solutions as HEP computing evolves.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

The large-area hybrid-optics CLAS12 RICH: First years of data-taking

The CLAS12 deep-inelastic scattering experiment at the upgraded 12 GeV continuous electron beam accelerator facility of Jefferson Lab conjugates luminosity and wide acceptance to study the 3D nucleon structure in the yet poorly explored valence region, and to perform precision measurements in hadron spectroscopy. A large area ring-imaging Cherenkov detector has been designed to achieve the required hadron identification in the momentum range from 3 GeV/c to 8 GeV/c, with the kaon rate about one order of magnitude lower than the rate of pions and protons. The adopted solution comprises aerogel radiator and composite mirrors in a novel hybrid optics design, where either direct or reflected light could be imaged in a high-packed and high-segmented photon detector. The first RICH module was assembled during the second half of 2017 and installed at the beginning of January 2018, in time for the start of the experiment. The second RICH module, planned with the goal to be ready for the beginning of the operation with polarized targets, has been timely built despite the complications caused by the pandemic crisis and successfully installed in June 2022. Finally, the detector performance is here discussed with emphasis on the operation and stability during the data-taking, calibration and alignment procedures, reconstruction and pattern recognition algorithms, and particle identification.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Rifle-like camera for long distance face recognition

An improved long-range facial recognition system is provided. The facial recognition system includes an integrated, rifle-like mounting platform with modular and interchangeable components, including a machine-vision camera, a facial recognition module, and an aiming scope. The mounting platform includes an elongated frame, a hand grip, a shoulder stock, and an underbody support attachment for an optional bipod. The mounting platform provides an intuitive, shoulder-operated support structure for stabilizing the machine-vision camera and reducing vibrations that otherwise inhibit long distance imaging.

Bolme, David S.↗
DOE OSTI2021

First 3D vector tracking of helium recoils for fast neutron measurements at Super$\mathrm{KEKB}$

Here we present results from the first deployment of novel, high definition, compact gas Time Projection Chambers (TPCs) with pixel chip readout as part of the BEAST II beam background measurement project at SuperKEKB. The TPCs provide detailed 3D imaging of ionization from neutron-induced nuclear recoils in a helium and carbon dioxide target gas mixture at standard temperature and pressure. We present the TPC performance and the neutron backgrounds observed during the initial stage of collider commissioning. We find excellent electron background rejection, leading to background-free nuclear recoil measurements above 50 keV ee , despite the extreme high-background environment. We measure an angular resolution better than 20° for recoil tracks longer than 1.7 mm, corresponding to an average ionization energy of approximately 100 keV ee . We also obtain the full 3D vector direction of helium recoils by utilizing charge profile measurements along the recoil axis, with a correct head/tail assignment efficiency of approximately 80%. With this performance, we present comparisons between measured and simulated event rates, recoil energy spectra, and directional distributions originating from beam–gas and Touschek beam losses at SuperKEKB. We utilize head/tail recognition to distinguish neutron components traveling with positive radial velocity in the Belle II coordinate system from those traveling in the opposite direction. Finally, we present a novel method of discriminating beam–gas interactions from Touschek beam losses that can eliminate the need for dedicated accelerator runs for background measurements. This method is still statistics-limited. However, future studies should be able to verify this method, which in turn could lead to neutron background analysis runs symbiotic with normal Belle II operation. The capabilities demonstrated here also suggest that high definition recoil imaging in gas TPCs is applicable to low energy, low-background experiments, such as directional dark matter searches.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Nowcasting Earthquakes: Imaging the Earthquake Cycle in California With Machine Learning

We propose a new machine learning-based method for nowcasting earthquakes to image the time-dependent earthquake cycle. The result is a timeseries that may correspond to the process of stress accumulation and release. The timeseries are constructed by using principal component analysis of regional seismicity. The patterns are found as eigenvectors of the cross-correlation matrix of a collection of seismicity timeseries in a coarse grained regional spatial grid (pattern recognition via unsupervised machine learning). The eigenvalues of this matrix represent the relative importance of the various eigenpatterns. Using the eigenvectors and eigenvalues, we compute the weighted correlation timeseries of the regional seismicity. This timeseries has the property that the weighted correlation generally decreases prior to major earthquakes in the region, and increases suddenly just after a major earthquake occurs. As in a previous paper, we find that this method produces a nowcasting timeseries that resembles the hypothesized regional stress accumulation and release process characterizing the earthquake cycle. We then address the problem of whether the timeseries contain information regarding future large earthquakes. For this, we compute a receiver operating characteristic and determine the decision thresholds for several future time periods of interest (optimization via supervised machine learning). We find that signals can be detected that can be used to characterize the information content of the timeseries. These signals may be useful in assessing present and near-future seismic hazards.

58 GEOSCIENCES↗
DOE OSTI2020

CENC PSN VOID FY20 Report

Maritime trade accounts for approximately 80 percent of international commerce. The high volume of vessels traversing domestic and international ports makes port areas prime targets for terrorism as well as illegal trafficking of drugs and arms (conventional or nuclear). Port security is therefore a worldwide concern affecting global economies, freedom of movement, and national security. However, extensive port monitoring is inherently complex and time consuming — making it truly viable only via an automated framework that can detect potential illicit activity and alert authorities in a timely manner. The development of image processing algorithms for this purpose requires access to large, labeled datasets that cover the breadth of targets of interest as well as the environments that they are observed within. Curated and labeled datasets of this nature are of enormous value to Sandia's Defense Nuclear Nonproliferation and National Security Program portfolios, as well as to Sandia's machine learning/automatic target recognition (ML/ATR) algorithm development and R&D communities. The goal of this project is to create a commercial satellite imagery dataset of labeled maritime vessels in port areas to support the development of ML/ATR algorithms for port security nonproliferation purposes. This dataset — Port Security Nonproliferation Vessel Overhead Imagery Dataset (PSN VOID) — has the potential to support a variety of other ancillary missions, such as maritime domain awareness, domestic and international security, drug interdiction, and weapons trafficking.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2020

DEEP CELLULAR RECURRENT NEURAL ARCHITECTURE FOR EFFICIENT MULTIDIMENSIONAL TIME-SERIES DATA PROCESSING

Efficient processing of time series data is a fundamental yet challenging problem in pattern recognition. Though recent developments in machine learning and deep learning have enabled remarkable improvements in processing large scale datasets in many application domains, most are designed and regulated to handle inputs that are static in time. Many real-world data, such as in biomedical, surveillance and security, financial, manufacturing and engineering applications, are rarely static in time, and demand models able to recognize patterns in both space and time. Current machine learning (ML) and deep learning (DL) models adapted for time series processing tend to grow in complexity and size to accommodate the additional dimensionality of time. Specifically, the biologically inspired learning based models known as artificial neural networks that have shown extraordinary success in pattern recognition, tend to grow prohibitively large and cumbersome in the presence of large scale multi-dimensional time series biomedical data such as EEG. Consequently, this work aims to develop representative ML and DL models for robust and efficient large scale time series processing. First, we design a novel ML pipeline with efficient feature engineering to process a large scale multi-channel scalp EEG dataset for automated detection of epileptic seizures. With the use of a sophisticated yet computationally efficient time-frequency analysis technique known as harmonic wavelet packet transform and an efficient self-similarity computation based on fractal dimension, we achieve state-of-the-art performance for automated seizure detection in EEG data. Subsequently, we investigate the development of a novel efficient deep recurrent learning model for large scale time series processing. For this, we first study the functionality and training of a biologically inspired neural network architecture known as cellular simultaneous recurrent neural network (CSRN). We obtain a generalization of this network for multiple topological image processing tasks and investigate the learning efficacy of the complex cellular architecture using several state-of-the?art training methods. Finally, we develop a novel deep cellular recurrent neural network (CDRNN) architecture based on the biologically inspired distributed processing used in CSRN for processing time series data. The proposed DCRNN leverages the cellular recurrent architecture to promote extensive weight sharing and efficient, individualized, synchronous processing of multi-source time series data. Experiments on a large scale multi-channel scalp EEG, and a machine fault detection dataset show that the proposed DCRNN offers state-of-the-art recognition performance while using substantially fewer trainable recurrent units.

Vidyaratne, Lasitha S.↗
DOE OSTI2022

The Facial Reconstruction of a Mesolithic Dog, Muge, Portugal

This paper presents the facial reconstruction of a Mesolithic dog whose skeleton was recovered from the Muge shell middens (Portugal) in the 19th century. We used the anatomical deformation approach based on a collection of computer tomography images as an attempt to reconstruct the Muge dog’s head appearance. We faced a few challenges due to the level of bone displacement and the absence of some cranium anatomical parts, as well as accurate information on soft tissue thickness for modern dogs. This multidisciplinary study combined anatomical, veterinary, zooarchaeological, artistic and graphic aspects to allow for the facial reconstruction of the Muge dog. Albeit an approximation, it confers a recognition to this prehistoric finding.

Moraes, Cicero (ORCID:0000000294790028)↗
DOE OSTI2022

Disentangling Alzheimer’s disease neurodegeneration from typical brain ageing using machine learning

Abstract Neuroimaging biomarkers that distinguish between changes due to typical brain ageing and Alzheimer’s disease are valuable for determining how much each contributes to cognitive decline. Supervised machine learning models can derive multivariate patterns of brain change related to the two processes, including the Spatial Patterns of Atrophy for Recognition of Alzheimer’s Disease (SPARE-AD) and of Brain Aging (SPARE-BA) scores investigated herein. However, the substantial overlap between brain regions affected in the two processes confounds measuring them independently. We present a methodology, and associated results, towards disentangling the two. T1-weighted MRI scans of 4054 participants (48–95 years) with Alzheimer’s disease, mild cognitive impairment (MCI), or cognitively normal (CN) diagnoses from the Imaging-based coordinate SysTem for AGIng and NeurodeGenerative diseases (iSTAGING) consortium were analysed. Multiple sets of SPARE scores were investigated, in order to probe imaging signatures of certain clinically or molecularly defined sub-cohorts. First, a subset of clinical Alzheimer’s disease patients (n = 718) and age- and sex-matched CN adults (n = 718) were selected based purely on clinical diagnoses to train SPARE-BA1 (regression of age using CN individuals) and SPARE-AD1 (classification of CN versus Alzheimer’s disease) models. Second, analogous groups were selected based on clinical and molecular markers to train SPARE-BA2 and SPARE-AD2 models: amyloid-positive Alzheimer’s disease continuum group (n = 718; consisting of amyloid-positive Alzheimer’s disease, amyloid-positive MCI, amyloid- and tau-positive CN individuals) and amyloid-negative CN group (n = 718). Finally, the combined group of the Alzheimer’s disease continuum and amyloid-negative CN individuals was used to train SPARE-BA3 model, with the intention to estimate brain age regardless of Alzheimer’s disease-related brain changes. The disentangled SPARE models, SPARE-AD2 and SPARE-BA3, derived brain patterns that were more specific to the two types of brain changes. The correlation between the SPARE-BA Gap (SPARE-BA minus chronological age) and SPARE-AD was significantly reduced after the decoupling (r = 0.56–0.06). The correlation of disentangled SPARE-AD was non-inferior to amyloid- and tau-related measurements and to the number of APOE ε4 alleles but was lower to Alzheimer’s disease-related psychometric test scores, suggesting the contribution of advanced brain ageing to the latter. The disentangled SPARE-BA was consistently less correlated with Alzheimer’s disease-related clinical, molecular and genetic variables. By employing conservative molecular diagnoses and introducing Alzheimer’s disease continuum cases to the SPARE-BA model training, we achieved more dissociable neuroanatomical biomarkers of typical brain ageing and Alzheimer’s disease.

Hwang, Gyujoon↗
DOE OSTI2021

Design, upgrade and characterization of the silicon photomultiplier front-end for the AMIGA detector at the Pierre Auger Observatory

AMIGA (Auger Muons and Infill for the Ground Array) is an upgrade of the Pierre Auger Observatory to complement the study of ultra-high-energy cosmic rays (UHECR) by measuring the muon content of extensive air showers (EAS). It consists of an array of 61 water Cherenkov detectors on a denser spacing in combination with underground scintillation detectors used for muon density measurement. Each detector is composed of three scintillation modules, with 10 m 2 detection area per module, buried at 2.3 m depth, resulting in a total detection area of 30 m 2 . Silicon photomultiplier sensors (SiPM) measure the amount of scintillation light generated by charged particles traversing the modules. In this paper, the design of the front-end electronics to process the signals of those SiPMs and test results from the laboratory and from the Pierre Auger Observatory are described. Compared to our previous prototype, the new electronics shows a higher performance, higher efficiency and lower power consumption, and it has a new acquisition system with increased dynamic range that allows measurements closer to the shower core. The new acquisition system is based on the measurement of the total charge signal that the muonic component of the cosmic ray shower generates in the detector.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

Forecast of Wildfire Potential Across California USA Using a Transformer

Wildfires are a major issue facing the United States, a matter further exacerbated by an ever-changing climate. In California alone, wildfires are responsible for billions of dollars in damages and take lives each year. Accurately predicting fire danger conditions allows preparation awareness before wildfires start. Transformers are a class of deep learning models designed to identify patterns in sequential datasets. In recent years, transformers have gained popularity through their impressive performance in natural language processing and other applications of signal recognition. This analysis demonstrates the ability of a transformer with a residual connection to forecast fire danger potential over the state of California. Wildland fire potential index (WFPI) maps collected from the US Geological Survey database from January 1st 2020 to December 31st 2023 were used to tune, train and evaluate the transformer. Meteorological inputs (provided by Daymet daily weather and climatological summaries), the normalized difference vegetation index (NDVI) (calculated from the Moderate Resolution Imaging Spectroradiometer (MODIS)), and outputs from the Scott and Burgman fire behavior fuel models (to characterize maps of fuel types), were used as inputs. Our results show that a transformer can effectively emulate the US Forest Service modeled WFPI maps of California USA for four week long forecasts over the month of July, 2023, with correlations ranging from 0.85 – 0.98.

Limber, Russell [ORNL]↗
DOE OSTI2025

Electromagnetic shower reconstruction in the ICARUS liquid argon time projection chamber detector

The ICARUS-T600 liquid argon time projection chamber (LArTPC) detector is taking data at shallow depth as the far detector of the Short Baseline Neutrino (SBN) program at Fermilab, to search for a possible sterile neutrino signal at $\Delta m^{2} \approx 1~\text{eV}^{2}$ with the Booster (BNB) and Main Injector (NuMI) neutrino beams at $\GeV{\sim 0.8}$ and $\GeV{\sim 2}$ average energies respectively. The LArTPC technology, developed by the ICARUS collaboration and now a standard in neutrino physics, offers impressive charged-particle imaging capabilities with $\sim1 \ \text{mm}$ spatial resolution, enabling efficient discrimination between track-like signatures (e.g., from muons, pions, and protons) and electromagnetic showers (from electrons and photons). Moreover, electron and photon signatures can be distinguished both with the calorimetric measurement of local energy depositions at the shower start and with the cm-scale conversion gap signature of photons. This contribution discusses event reconstruction at ICARUS focusing on Pandora, a multi-algorithm pattern recognition software widely used in LArTPC experiments. Over a hundred Pandora algorithms and tools are used to reconstruct cosmic rays and neutrino interactions in the ICARUS detector. Recent developments have focused on the reconstruction of electromagnetic shower signatures, crucial to ensure a robust and efficient reconstruction of charged-current $\nu_e$ interactions, which serve as a key signature of sterile neutrino oscillations at SBN. In this contribution, recent improvements to the reconstruction are discussed, focusing on the discrimination between tracks and electromagnetic showers using neutrino simulations and data.

Triozzi, Riccardo [Padua U.; INFN, Padua] (ORCID:0↗
DOE OSTI2025

Using Eye Tracking to Elucidate the Mechanisms Underlying Stimulation-Enhanced Visual Target Detection

Transcranial direct current stimulation (tDCS) is a noninvasive form of brain stimulation that involves passing a weak electrical current between electrodes on the scalp to modulate underlying neural tissue. TDCS has been shown to modulate cognition in a variety of domains, including memory, attention, and visual processing. Prior work from our laboratory has shown positive effects of tDCS on learning to detect target objects hidden in complex naturalistic visual scenes and learn rules for categorizing images, though the mechanism for these benefits remains unknown. One possibility is that tDCS optimizes visual search by modulating visual attention or via the reduction in search errors. One method of quantifying visual attention is to use eye tracking to record search patterns to determine if and how visual search is adjusted under verum stimulation conditions. Eye tracking data allows classification of errors into error types, including sampling errors (failing to look in the relevant region), recognition errors (looking at the critical portion of a scene, but failing to recognize it as such as evidenced by visual fixation), and decision-making errors (fixating on the relevant portion of a scene, but making the wrong determination). Our results indicate that the benefit tDCS confers on visual search for targets stems from the reduction in decision-making errors when targets are present (Cohen’s d = 0.86). Also reported is a replication of previous findings showing a tDCS-dependent improvement in learning this task, learning score (Cohen’s d = 0.88); d’ (Cohen’s d = 1.00). This provides support for moving tDCS into the application space by pairing it with analysts who are concerned with the type of search error that is corrected via stimulation.

attention↗