Search NASA⌕ Search

SEARCH · Search NASA

Results for “Image segmentation”

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 487 records · Page 27
All sourcesNASA NTRSDOE OSTI
NASA NTRS1984

LANDSAT-D program. Volume 2: Ground segment

Raw digital data, as received from the LANDSAT spacecraft, cannot generate images that meet specifications. Radiometric corrections must be made to compensate for aging and for differences in sensitivity among the instrument sensors. Geometric corrections must be made to compensate for off-nadir look angle, and to calculate spacecraft drift from its prescribed path. Corrections must also be made for look-angle jitter caused by vibrations induced by spacecraft equipment. The major components of the LANDSAT ground segment and their functions are discussed.

Source record↗
NASA NTRS2006

Phase Sensor for Aligning a Segmented Telescope Mirror

A phase sensor has been developed for use in aligning a segmented telescope mirror to within a fraction of a wavelength in piston. (As used here, piston signifies displacement of a mirror segment along the optical axis of the telescope.) Such precise alignment is necessary in order to realize the full benefit of the large aperture achievable through segmentation. This phase sensor is achromatic. It is based on two-wavelength shearing interferometry, and can be modified to utilize an extended or broad-band (e.g., white) light source. The sensor optics include a ruled diffraction grating and an imaging lens. The sensor can measure the piston shift between segments as well as aberrations of the segments. It can measure the surface error of an individual segment, making it possible to compensate for the error with optimal amount(s) of piston and/or tilt. The precise capture range of the sensor depends partly on the telescope design; the largest relative piston shifts measurable by use of this sensor are of the order of 100 m. The accuracy of the sensor also depends partly on the telescope design; in general, the accuracy is sufficient to enable alignment to within approximately half a wavelength. The interferometric image is digitized and processed by a simple algorithm in real time, and the output of the algorithm can be used to maintain alignment in real time, even in the presence of atmospheric turbulence. The sensor is robust. Through calibration, it can be made insensitive to (and, hence, tolerant of) misalignments and aberrations of its own optics, most aberrations of the telescope as a whole (in contradistinction to aberrations of individual segments), and most aberrations introduced by atmospheric turbulence

Stahl, Philip↗
NASA NTRSDate not supplied

Kalman Filtering and RTS Smoothing for Arc-Jet Sample Edge Tracking

Accurate and temporally consistent measurements of test-article recession are required to characterize material response during arc-jet ablation experiments. However, image-based boundary measurements are often affected by segmentation noise, brightness variations, and frame-to-frame variability. This work extends arcjetCV [1] with filtering methods for tracking the evolving boundaries of hemispherical and ISO-Q test articles. Two boundary-tracking approaches based on Kalman filtering [2] were implemented. The first applies a point-wise Kalman filter followed by a Rauch–Tung–Striebel smoother [3] to individual boundary-point locations. The second applies the same filtering and smoothing framework to a reduced set of geometry-dependent shape parameters. The point-wise method reduces local frame-to-frame fluctuations while preserving spatial details along the detected boundary. The shape-parameter method provides a compact and geometrically constrained estimate of the sample contour. Figure 1 illustrates the two approaches for a hemispherical test article. Both methods improve temporal consistency and support more robust estimation of surface recession from arc-jet imagery. The two filtering approaches provide complementary representations of boundary evolution and improve the reliability of image-based recession measurements during arc-jet experiments.

recession measurement↗
NASA NTRS2012

Hyperspectral Feature Detection Onboard the Earth Observing One Spacecraft using Superpixel Segmentation and Endmember Extraction

We present a demonstration of onboard hyperspectral image processing with the potential to reduce mission downlink requirements. The system detects spectral endmembers and then uses them to map units of surface material. This summarizes the content of the scene, reveals spectral anomalies warranting fast response, and reduces data volume by two orders of magnitude. We have integrated this system into the Autonomous Science craft Experiment for operational use onboard the Earth Observing One (EO-1) Spacecraft. The system does not require prior knowledge about spectra of interest. We report on a series of trial overflights in which identical spacecraft commands are effective for autonomous spectral discovery and mapping for varied target features, scenes and imaging conditions.

Superpixel Endmember Detection↗
NASA NTRSDate not supplied

Laboratory Demonstration of High Contrast with the PIAACMC Coronagraph on an Obstructed and Segmented Aperture

Coronagraphs (together with starshades) are an important tool to directly image and characterize exoplanets, and enable the search for biomarkers in reflected light on potentially habitable worlds. Their performance and efficiency has steadily been improving over the past several decades, but has not yet reached performance limits. In particular, the expected exoplanet yield for missions such as the Astro2020-recommended “IR/O/UV Flagship” can still be improved by factors of at least 2-3, simply by continued improvements in coronagraph performance, before they plateau due to physics limits. One possible architecture that can enable at least a part of this improvement is the Phase-Induced Amplitude Apodization Complex Mask Coronagraph (PIAACMC). Itoffersthe advantages of high throughput, small inner working angle (IWA),and almost noloss inPSF sharpness, and natively supports obstructed and segmented apertures, which is essential to the Astro2020 “IR/O/UV Flagship”mission. Historically, key disadvantages of PIAA have been poor tolerance to stellar angular sizes and maturity, but latest designs and demonstrations have made significant strides in this respect.In this paper, we present the current status and overview of our program to mature the PIAACMC technology. We first review PIAACMC designs for LUVOIR-A and B, which resultedin improved expected yield of Exo-Earths relative tothe baselines for both mission concepts. In particular, for LUVOIR-B, the yield improves from 28 to 42due to improvedtolerance to stellar angular size in our design. The improvement in yield is marginal for LUVOIR-A, but our design enables detecting planets around smaller diameter stars (nearby K-dwarfs and/or farther FG stars). We also describe our modeling and vacuum demonstrationsfor the LUVOIR-A aperture(which is more challengingthan LUVOIR-B due to the central obstruction).The demonstration included a LUVOUR-A pupil mask, an on-axis PIAA set of 2 mirrors with central holes, a Boston Micromachines DM, a patterned CMC mask, Lyot stop, and supporting masks and optics. Demonstrations were conducted at JPL’s High Contrast Imaging Testbed (HCIT) using several characterization andwavefront control techniques (primarily standard EFC, several experimental techniques were attempted, as well as speckle nulling). So far, our results include 1.9e-8 raw contrast in 10% broadband light between 3.5 and 8 l/D; 4.1e-8 and 1.6e-9 coherent contrasts in monochromatic light between 2-4 and 4-8 l/D, respectively. We also present measurements and analysis of sensitivity to tip/tilt jitter and stellar angular size. Finally, we compare our test results to models, present an analysis of our limiting factors, and explorefuture prospectsfor improvement based on validated models.

Coronagraph↗
DOE OSTI2024

Automating the detection of hydrological barriers and fragmentation in wetlands using deep learning and InSAR

The loss of hydrological connectivity and fragmentation of natural wetlands is a widespread driver of wetland degradation. Understanding where and how natural connectivity is impaired is essential for managing, protecting and remediating these ecosystems. Wetland Interferometric Synthetic Aperture Radar (Wetland InSAR) can provide information on surface flow orientation in wetlands at a high spatial resolution, which can be used for barrier detection. However, the broad application of this approach is constrained by the labour-intensive manual delineation of barriers based on mapped water levels. This study presents the first deep learning-based methodology for the automated detection of hydrological barriers. We trained a deep convolutional network to segment edge features of hydrological barriers in 25 image pairs captured by ALOS PALSAR-1 L-Band InSAR between 2006 and 2011. The training dataset consists of manually labelled and delineated barriers showing abrupt changes in water surface elevation and wrapped interferograms with high coherence. We tested this method across three wetland sites: the Everglades and southern Louisiana wetlands (United States) and the Cienaga de Zapata (Cuba). Across these sites, the convolutional network detected hydrological barriers with up to 84% accuracy. The model performed particularly well for linear hydrological barriers such as roads, dikes, and channels. Notably, some barriers impede flow only seasonally, appearing during low water levels and disappearing when water levels rise. Our automated approach to detecting and assessing wetland hydrologic connectivity can be applied more broadly to support the effective management of fragmented wetland ecosystems.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1990

Eastern Aphrodite Terra on Venus - Characteristics, structure, and mode of origin

Altimetric and radar-image cross-strike discontinuities (CSDs) are found to cut across the nearly east-west strike of Eastern Aphrodite (EA) Terra, dividing or segmenting it into offset domains. Each of these domains is characterized by a discontinuous central chasma or trough. These features have been mapped in the altimetric and radar image data of EA on the basis of abrupt termination of riftlike central chasma, offset, and segmentation of the center highlands, and radar image discontinuities previously mapped in Western Aphrodite in terms of length, orientation, and influence on the central highlands and adjacent lowlands. It is concluded that EA displays a variety of geological and geophysical characteristics similar to those occuring at divergent boundaries. The surface map characteristics are the result of a great horizontal motion of the crust diverging from a well-defined linear region, a process that can occur whether the surface is part of a thick and strong plate detached from mantle convection or part of a thin and weak layer attached to a deep convective limb.

Crumpler, L. S.↗
NASA NTRS2006

NASA Tech Briefs, September 2006

Topics covered include: Improving Thermomechanical Properties of SiC/SiC Composites; Aerogel/Particle Composites for Thermoelectric Devices; Patches for Repairing Ceramics and Ceramic- Matrix Composites; Lower-Conductivity Ceramic Materials for Thermal-Barrier Coatings; An Alternative for Emergency Preemption of Traffic Lights; Vehicle Transponder for Preemption of Traffic Lights; Automated Announcements of Approaching Emergency Vehicles; Intersection Monitor for Traffic-Light-Preemption System; Full-Duplex Digital Communication on a Single Laser Beam; Stabilizing Microwave Frequency of a Photonic Oscillator; Microwave Oscillators Based on Nonlinear WGM Resonators; Pointing Reference Scheme for Free-Space Optical Communications Systems; High-Level Performance Modeling of SAR Systems; Spectral Analysis Tool 6.2 for Windows; Multi-Platform Avionics Simulator; Silicon-Based Optical Modulator with Ferroelectric Layer; Multiplexing Transducers Based on Tunnel-Diode Oscillators; Scheduling with Automated Resolution of Conflicts; Symbolic Constraint Maintenance Grid; Discerning Trends in Performance Across Multiple Events; Magnetic Field Solver; Computing for Aiming a Spaceborne Bistatic- Radar Transmitter; 4-Vinyl-1,3-Dioxolane-2-One as an Additive for Li-Ion Cells; Probabilistic Prediction of Lifetimes of Ceramic Parts; STRANAL-PMC Version 2.0; Micromechanics and Piezo Enhancements of HyperSizer; Single-Phase Rare-Earth Oxide/Aluminum Oxide Glasses; Tilt/Tip/Piston Manipulator with Base-Mounted Actuators; Measurement of Model Noise in a Hard-Wall Wind Tunnel; Loci-STREAM Version 0.9; The Synergistic Engineering Environment; Reconfigurable Software for Controlling Formation Flying; More About the Tetrahedral Unstructured Software System; Computing Flows Using Chimera and Unstructured Grids; Avoiding Obstructions in Aiming a High-Gain Antenna; Analyzing Aeroelastic Stability of a Tilt-Rotor Aircraft; Tracking Positions and Attitudes of Mars Rovers; Stochastic Evolutionary Algorithms for Planning Robot Paths; Compressible Flow Toolbox; Rapid Aeroelastic Analysis of Blade Flutter in Turbomachines; General Flow-Solver Code for Turbomachinery Applications; Code for Multiblock CFD and Heat-Transfer Computations; Rotating-Pump Design Code; Covering a Crucible with Metal Containing Channels; Repairing Fractured Bones by Use of Bioabsorbable Composites; Kalman Filter for Calibrating a Telescope Focal Plane; Electronic Absolute Cartesian Autocollimator; Fiber-Optic Gratings for Lidar Measurements of Water Vapor; Simulating Responses of Gravitational-Wave Instrumentation; SOFTC: A Software Correlator for VLBI; Progress in Computational Simulation of Earthquakes; Database of Properties of Meteors; Computing Spacecraft Solar-Cell Damage by Charged Particles; Thermal Model of a Current-Carrying Wire in a Vacuum; Program for Analyzing Flows in a Complex Network; Program Predicts Performance of Optical Parametric Oscillators; Processing TES Level-1B Data; Automated Camera Calibration; Tracking the Martian CO2 Polar Ice Caps in Infrared Images; Processing TES Level-2 Data; SmaggIce Version 1.8; Solving the Swath Segment Selection Problem; The Spatial Standard Observer; Less-Complex Method of Classifying MPSK; Improvement in Recursive Hierarchical Segmentation of Data; Using Heaps in Recursive Hierarchical Segmentation of Data; Tool for Statistical Analysis and Display of Landing Sites; Automated Assignment of Proposals to Reviewers; Array-Pattern-Match Compiler for Opportunistic Data Analysis; Pre-Processor for Compression of Multispectral Image Data; Compressing Image Data While Limiting the Effects of Data Losses; Flight Operations Analysis Tool; Improvement in Visual Target Tracking for a Mobile Robot; Software for Simulating Air Traffic; Automated Vectorization of Decision-Based Algorithms; Grayscale Optical Correlator Workbench; "One-Stop Shopping" for Ocean Remote-Sensing and Model Data; State Analysis Database Tool; Generating CAHV and CAHVOmages with Shadows in ROAMS; Improving UDP/IP Transmission Without Increasing Congestion; FORTRAN Versions of Reformulated HFGMC Codes; Program for Editing Spacecraft Command Sequences; Flight-Tested Prototype of BEAM Software; Mission Scenario Development Workbench; Marsviewer; Tool for Analysis and Reduction of Scientific Data; ASPEN Version 3.0; Secure Display of Space-Exploration Images; Digital Front End for Wide-Band VLBI Science Receiver; Multifunctional Tanks for Spacecraft; Lightweight, Segmented, Mostly Silicon Telescope Mirror; Assistant for Analyzing Tropical-Rain-Mapping Radar Data; and Anion-Intercalating Cathodes for High-Energy- Density Cells.

Source record↗
NASA NTRS1995

Algorithms for detection of objects in image sequences captured from an airborne imaging system

This research was initiated as a part of the effort at the NASA Ames Research Center to design a computer vision based system that can enhance the safety of navigation by aiding the pilots in detecting various obstacles on the runway during critical section of the flight such as a landing maneuver. The primary goal is the development of algorithms for detection of moving objects from a sequence of images obtained from an on-board video camera. Image regions corresponding to the independently moving objects are segmented from the background by applying constraint filtering on the optical flow computed from the initial few frames of the sequence. These detected regions are tracked over subsequent frames using a model based tracking algorithm. Position and velocity of the moving objects in the world coordinate is estimated using an extended Kalman filter. The algorithms are tested using the NASA line image sequence with six static trucks and a simulated moving truck and experimental results are described. Various limitations of the currently implemented version of the above algorithm are identified and possible solutions to build a practical working system are investigated.

Kasturi, Rangachar↗
DOE OSTI2024

Uncertainty-aware particle segmentation for electron microscopy at varied length scales

Electron microscopy is indispensable for examining the morphology and composition of solid materials at the sub-micron scale. To study the powder samples that are widely used in materials development, scanning electron microscopes (SEMs) are increasingly used at the laboratory scale to generate large datasets with hundreds of images. Parsing these images to identify distinct particles and determine their morphology requires careful analysis, and automating this process remains challenging. In this work, we enhance the Mask R-CNN architecture to develop a method for automated segmentation of particles in SEM images. We address several challenges inherent to measurements, such as image blur and particle agglomeration. Moreover, our method accounts for prediction uncertainty when such issues prevent accurate segmentation of a particle. Recognizing that disparate length scales are often present in large datasets, we use this framework to create two models that are separately trained to handle images obtained at low or high magnification. By testing these models on a variety of inorganic samples, our approach to particle segmentation surpasses an established automated segmentation method and yields comparable results to the predictions of three domain experts, revealing comparable accuracy while requiring a fraction of the time. These findings highlight the potential of deep learning in advancing autonomous workflows for materials characterization.

36 MATERIALS SCIENCE↗
NASA NTRS2004

Next Generation Geostationary Operational Environmental Satellite (GOES-R Series): A Space Segment Overview

The next-generation National Oceanic and Atmospheric Administration (NOAA) Geostationary Operational Environmental Satellite (GOES-R series) is currently being developed by NOAA in cooperation with the National Aeronautics and Space Administration (NASA). The GOES-R series satellites represents a significant improvement in spatial, temporal, and spectral observations (several orders of magnitude) over the capabilities of the currently operational GOES-1 series and the about to be launched GOES-N series satellite. The GOES-R series will incorporate technically advanced third-generation instruments and spacecraft enhancements to meet evolving observational requirements of forecasting for the era 2012-2025. The GOES-R instrument complement being developed includes a Advanced Baseline Imager (ABI), a Hyperspectral Environmental Suite (HES), a GEO Lighting Mapper (GLM), a Solar Imaging Suite (SIS) and a Space Environment In-Situ Suite (SEISS). Also, candidates for a number of GOES-R Pre-Planned Product Improvements (P(sup 3)Is) includes a Geo microwave Sounder, a Coronograph, a Hyperspectral Imager, and a Solar Irradiance Sensor. Currently, the GOES-R Space Segment architecture is being evaluated as part of a GOES-R system end-to-end architecture study. The GOES-R notional baseline architecture is a constellation of two satellites (A-sat and B-sat) each nominally located at 75 degrees west longitude and at 135 degrees west longitude at geostationary altitude, 0 degrees inclination. The primary mission of the A-sat is to provide imaging from the ABI. The A-sat will also contain the SIS and the GLM. The primary mission of the B-sat is to provide sounding of the hemispherical disk of the earth from the HES. The B-sat also contains the SEISS. Both satellites have mesoscale capabilities for severe weather sounding or imaging. This paper overviews the GOES-R Space Segment development including satellite constellation trade-off, improvements and differences between the current and future instrument and spacecraft capabilities, and technology infusion.

Krimchansky, Alexander↗
NASA NTRS2006

Effects of diffraction and static wavefront errors on high-contrast imaging from the Thirty Meter Telescope

High-contrast imaging, particularly direct detection of extrasolar planets, is a major science driver for the next generation of extremely large telescopes such as the segmented Thirty Meter Telescope. This goal requires more than merely diffraction-limited imaging, but also attention to residual scattered light from wavefront errors and diffraction effects at the contrast level of 10-8-10-9. Using a wave-optics simulation of adaptive optics and a diffraction suppression system we investigate diffraction from the segmentation geometry, intersegment gaps, obscuration by the secondary mirror and its supports. We find that the large obscurations pose a greater challenge than the much smaller segment gaps. In addition the impact of wavefront errors from the primary mirror, including segment alignment and figure errors, are analyzed. Segment-to-segment reflectivity variations and residual segment figure error will be the dominant error contributors from the primary mirror. Strategies to mitigate these errors are discussed.

segmented mirrors↗
DOE OSTI2026

NanoPSD: A software for automatic detection of Nano-Particle Shape Distribution in electron microscopy images

Accurate quantification of the size and morphology of nanoparticles from electron microscopy (EM) images is essential to understand growth mechanisms, surface reactivity, and functional behavior in nanoscale materials. Manual analysis remains slow, subjective, and difficult to reproduce in large datasets. We introduce NanoPSD (Nano-Particle Shape Distribution), an open-source and fully automated framework for quantitative particle detection and morphology analysis from EM images. NanoPSD integrates adaptive contrast enhancement, polarity-agnostic scale-bar detection, Optical Character Recognition (OCR)-based calibration, and classical segmentation via Otsu thresholding with morphological refinement. Particle contours are used to extract geometric descriptors, including equivalent circular diameter, aspect ratio, circularity, and solidity, enabling automated classification into spherical, rod-like, and aggregate morphologies. The framework supports both single-image and batch processing, generating publication-quality visualizations, LaTeX-ready tables, and structured comma-separated values (CSV) datasets. As a demonstration, we applied NanoPSD to plasma-synthesized nanoparticle samples diagnosed via transmission electron microscopy (TEM). The code produced statistically robust size and morphology distributions spanning a few to tens of nanometers with minimal user supervision. The pipeline demonstrates high reproducibility and scalability, processing large image collections with consistent calibration and output formatting. Its modular design enables seamless integration of future deep-learning-based segmentation models, providing a pathway toward intelligent, data-driven electron microscopy analysis.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Leveraging unlabeled SEM datasets with self-supervised learning for enhanced particle segmentation

Scanning Electron Microscopes (SEMs) are widely used in experimental science laboratories, often requiring cumbersome and repetitive user analysis. Automating SEM image analysis processes is highly desirable to address this challenge. In particle sample analysis, Machine Learning (ML) has emerged as the most effective approach for particle segmentation. However, the time-intensive process of manually annotating thousands of SEM images limits the applicability of supervised learning approaches. Self-Supervised Learning (SSL) offers a promising alternative by enabling knowledge extraction from raw, unlabeled data. This study presents a framework for evaluating SSL techniques in SEM image analysis, focusing on novel methods leveraging the ConvNeXtV2 architecture for particle detection. A dataset comprising 25,000 SEM images is curated to benchmark these proposed SSL methods. The results demonstrate that ConvNeXtV2 models, with varying parameter counts, consistently outperform other techniques in particle detection across different length scales, achieving up to a 34% reduction in relative error compared to established SSL methods. Furthermore, an ablation study explores the relationship between dataset size and SSL performance, providing actionable insights for practitioners regarding model selection and resource efficiency. This research advances the integration of SSL into autonomous analysis pipelines and supports its application in accelerating materials science discovery.

Rettenberger, Luca↗
DOE OSTI2024

Electric field enhanced diffusion welding of alloy 617: Microstructural characteristics and mechanical properties

This study investigated the microstructural characteristics and mechanical behavior of diffusion welded nickel-based Alloy 617 obtained by electric field-assisted sintering (EFAS) using various parameters. The interfacial microstructure exhibited different characteristics including good grain boundary (GB) migration across the interface in the samples diffusion-welded at 1100 °C and a flat interface in the samples joined at 1000 °C and 1050 °C. The interface consisted of fine Al 2 O 3 oxides, while precipitation of interfacial M 23 C 6 carbides was not observed. Grain boundaries migrated across the Al 2 O 3 oxides, leaving these oxides within the grains. Graded grain size was observed, with grain coarsening being more significant near the sample surface due to the temperature gradient induced by EFAS. Tensile testing revealed that the specimens fractured in the matrix away from the interface, indicting strong diffusion-welded joints. Further, the peak tensile strength of 807 MPa was obtained in the samples welded at 1000 °C due to minimal grain growth. The materials obtained at 1100 °C exhibited reduced tensile strength but improved ductility. Strain maps revealed by digital image correlation showed alternating high and low strain segments in the samples produced at 1000 °C and 1050 °C, indicating that the flat interfaces with no GB migration were less ductile compared to the matrix. A greater strain uniformity was observed along the bond interfaces with improved GB migration. The hardness reduced near the sample surfaces due to enlarged grains induced by temperature gradient. This study demonstrates that GB migration and enhanced mechanical strength can be achieved in diffusion-welded Alloy 617.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Measuring thermal curing induced shrinkage of material extrusion based additive manufacturing silicone elastomer lattices by X-ray computed tomography

Thermal curing induces shrinkage in material extrusion based additive manufacturing silicone elastomer samples, resulting in discrepancies between as printed and final geometries. Knowing the extent to which the samples change in shape and size allows us to make appropriate modifications to the printing design to better control the geometry of the samples. We present an X-ray computed tomography (CT) based approach to determine filament-level shrinkage due to thermal curing of silicone elastomer samples printed with direct ink writing (DIW). The approach relies on custom-designed build plates that are resistant to the elevated curing temperatures and that have sufficiently distinct X-ray absorption characteristics from the silicone elastomer to ensure adequate segmentation of the latter in X-ray imaging data. We implement the approach to evaluate shrinkage in DIW ‘log pile’ samples with three distinct strand arrangements and demonstrate of how filament-level information can be extracted from the X-ray CT data.

Additive manufacturing↗
DOE OSTI2024

Bragg Spot Finder (BSF): a new machine-learning-aided approach to deal with spot finding for rapidly filtering diffraction pattern images

Macromolecular crystallography contributes significantly to understanding diseases and, more importantly, how to treat them by providing atomic resolution 3D structures of proteins. This is achieved by collecting X-ray diffraction images of protein crystals from important biological pathways. Spotfinders are used to detect the presence of crystals with usable data, and the spots from such crystals are the primary data used to solve the relevant structures. Having fast and accurate spot finding is essential, but recent advances in synchrotron beamlines used to generate X-ray diffraction images have brought us to the limits of what the best existing spotfinders can do. This bottleneck must be removed so spotfinder software can keep pace with the X-ray beamline hardware improvements and be able to see the weak or diffuse spots required to solve the most challenging problems encountered when working with diffraction images. In this paper, we first present Bragg Spot Detection (BSD), a large benchmark Bragg spot image dataset that contains 304 images with more than 66 000 spots. We then discuss the open source extensible U-Net-based spotfinder Bragg Spot Finder (BSF), with image pre-processing, a U-Net segmentation backbone, and post-processing that includes artifact removal and watershed segmentation. Finally, we perform experiments on the BSD benchmark and obtain results that are (in terms of accuracy) comparable to or better than those obtained with two popular spotfinder software packages ( Dozor and DIALS ), demonstrating that this is an appropriate framework to support future extensions and improvements.

36 MATERIALS SCIENCE↗
DOE OSTI2025

DeepAndes: A Self-Supervised Vision Foundation Model for Multispectral Remote Sensing Imagery of the Andes

By mapping sites at large scales usingremotely sensed data, archaeologists can generate unique insights into long-term demographic trends, interregional social networks, and human adaptations in the past. Remote sensing surveys complement field-based approaches, and their reach can be especially great when combined with deep learning and computer vision techniques. However, conventional supervised deep learning methods face challenges in annotating fine-grained archaeological features at scale. In addition, while recent vision foundation models have shown remarkable success in learning large-scale remote sensing data with minimal annotations, most off-the-shelf solutions are designed for RGB images rather than multispectral satellite imagery, such as the eight-band data used in our study. In this article, we introduce DeepAndes, a transformer-based vision foundation model trained on three million multispectral satellite images, specifically tailored for Andean archaeology. DeepAndes incorporates a customized DINOv2 self-supervised learning algorithm optimized for eight-band multispectral imagery, marking the first foundation model designed explicitly for the Andes region. We evaluate its image understanding performance through imbalanced image classification, image instance retrieval, and pixel-level semantic segmentation tasks. Our experiments show that DeepAndes achieves superior F1 scores, mean average precision, and Dice scores in few-shot learning scenarios, significantly outperforming models trained from scratch or pretrained on smaller datasets. This underscores the effectiveness of large-scale self-supervised pretraining in archaeological remote sensing.

Guo, Junlin [Vanderbilt Univ., Nashville, TN (Unit↗