Search NASA⌕ Search

SEARCH · Search NASA

Results for “Selective classification”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2014

Electronic nose with a new feature reduction method and a multi-linear classifier for Chinese liquor classification

An electronic nose (e-nose) was designed to classify Chinese liquors of the same aroma style. A new method of feature reduction which combined feature selection with feature extraction was proposed. Feature selection method used 8 feature-selection algorithms based on information theory and reduced the dimension of the feature space to 41. Kernel entropy component analysis was introduced into the e-nose system as a feature extraction method and the dimension of feature space was reduced to 12. Classification of Chinese liquors was performed by using back propagation artificial neural network (BP-ANN), linear discrimination analysis (LDA), and a multi-linear classifier. The classification rate of the multi-linear classifier was 97.22%, which was higher than LDA and BP-ANN. Finally the classification of Chinese liquors according to their raw materials and geographical origins was performed using the proposed multi-linear classifier and classification rate was 98.75% and 100%, respectively.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Merger Signatures are Common, but not Universal, in Massive, Recently Quenched Galaxies at z ~ 0.7

We present visual classifications of merger-induced tidal disturbances in 143 M* ~ 10 11 M ⊙ post-starburst galaxies at z ~ 0.7 identified in the $SQuIGG\vec{L}E$ Sample. This sample spectroscopically selects galaxies from the Sloan Digital Sky Survey that have stopped their primary epoch of star formation within the past ~500 Myr. Visual classifications are performed on Hyper Suprime-Cam imaging. We compare to a control sample of mass- and redshift-matched star-forming and quiescent galaxies from the Large Early Galaxy Census and find that post-starburst galaxies are more likely to be classified as disturbed than either category. This corresponds to a factor of $3.6$$^{+2.9}_{-1.3}$ times the disturbance rate of older quiescent galaxies and $2.1$$^{+1.9}_{-.73}$ times the disturbance rate of star-forming galaxies. Assuming tidal features persist for ≲500 Myr, this suggests merging is coincident with quenching in a significant fraction of these post-starbursts. Galaxies with tidal disturbances are younger on average than undisturbed post-starburst galaxies in our sample, suggesting tidal features from a major merger may have faded over time. This may be exacerbated by the fact that, on average, the undisturbed subset is fainter, rendering low-surface-brightness tidal features harder to identify. However, the presence of 10 young (≲150 Myr since quenching) undisturbed galaxies suggests that major mergers are not the only fast physical mechanism that shut down the primary epoch of star formation in massive galaxies at intermediate redshift.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Forest structure and solar-induced fluorescence across intact and degraded forests in the Amazon

Tropical forest degradation (e.g., anthropogenic disturbances such as selective logging and fires) alters forest structure and function and influences the forest's carbon sink. In this study, we explored structure-function relationships across a variety of degradation levels in the southern Brazilian Amazon by 1) investigating how forest structural properties vary as a function of degradation history using airborne lidar data; 2) assessing the effects of degradation on solar-induced chlorophyll fluorescence (SIF) seasonality using TROPOMI data; and 3) quantifying the contribution of structural variables to SIF using multiple regression models with stepwise selection of lidar metrics. Forest degradation history was obtained through Landsat time-series classification. We found that fire, logging, and time since disturbance were major determinants of forest structure, and that forests affected by fires experienced larger variability in leaf area index (LAI), canopy height and vertical structure relative to logged and intact forests. Moreover, only recently burned forests showed significantly depressed SIF during the dry season compared to intact forests. Canopy height and the vertical distribution of foliage were the best predictors of SIF. Unexpectedly, we found that wet-season SIF was higher in active regenerating forests (~ 4 years after fires or logging) compared with intact forests, despite lower LAI. Furthermore, our findings help to elucidate the mechanisms of carbon accumulation in anthropogenically disturbed tropical forests and indicate that they can capture large amounts of carbon while recovering.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2013

DA WHITE DWARFS OBSERVED IN THE LAMOST PILOT SURVEY

A total of {approx}640, 000 objects from the LAMOST pilot survey have been publicly released. In this work, we present a catalog of DA white dwarfs (DAWDs) from the entire pilot survey. We outline a new algorithm for the selection of white dwarfs (WDs) by fitting Sersic profiles to the Balmer H{beta}, H{gamma}, and H{delta} lines of the spectra, and calculating the equivalent width of the Ca II K line. Two thousand nine hundred sixty-four candidates are selected by constraining the fitting parameters and the equivalent width of the Ca II K line. All the spectra of candidates are visually inspected. We identify 230 DAWDs (59 of which are already included in the Villanova and SDSS WD catalogs), 20 of which are DAWDs with non-degenerate companions. In addition, 128 candidates are classified as DAWDs/subdwarfs, which means the classifications are ambiguous. The result is consistent with the expected DAWD number estimated based on the LEGUE target selection algorithm.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

Organic Chemical Attribution Signatures for the Sourcing of a Mustard Agent and Its Starting Materials

Chemical attribution signatures (CAS) are being investigated for the sourcing of chemical warfare (CW) agents and their starting materials that may be implicated in chemical attacks or CW proliferation. The work reported here demonstrates for the first time trace impurities produced during the synthesis of tris(2-chloroethyl)amine (HN3) that point to specific reagent stocks used in the synthesis of this CW agent. Thirty batches of HN3 were synthesized using different combinations of commercial stocks of triethanolamine (TEA), thionyl chloride, chloroform, and acetone. The HN3 batches and reagent stocks were then analyzed for impurities by gas chromatography/mass spectrometry. Reaction-produced impurities indicative of specific TEA and chloroform stocks were exclusively discovered in HN3 batches made with those reagent stocks. In addition, some reagent impurities were found in the HN3 batches that were presumably not altered during synthesis and believed to be indicative of reagent type regardless of stock. Supervised classification using partial least squares discriminant analysis (PLSDA) on the impurity profiles of chloroform samples from seven stocks resulted in an average classification error by cross-validation of 2.4%. A classification error of zero was obtained using the seven-stock PLSDA model on a validation set of samples from an arbitrarily selected chloroform stock. In a separate analysis, all samples from two of seven chloroform stocks that were purposely not modeled had their samples matched to a chloroform stock rather than assigned a “no class” classification.

chemical forensics↗
DOE OSTI2025

JGI-Trichoderma v1.0

There is a series of Python and bash scripts to parse genomics datasets used to evaluate the coevolution of gene families and the feature importance of gene families using an SVM classifier. - Cover analysis: takes a list of single-copy genes in a set of genomes, aligns and builds the gene trees to determine if two gene families have a signature of covariation with one another. It parses the files to run phykit cover script described here: https://jlsteenwyk.com/PhyKIT/usage/index.html - SVM-classifier: This Python script is an SVM-based genomic classifier designed for biological data analysis. It combines machine learning with feature selection to identify important genomic markers and classify biological samples. Core Functionality: The script uses Support Vector Machines from scikit-learn to classify genomic data, incorporating SelectKBest for automated feature selection and leave-one-out cross-validation for performance assessment. It operates in multiple modes: feature ranking, optimal combination discovery, and sample prediction. Primary Applications: Genomic sample classification and biomarker discovery Feature importance analysis in high-dimensional biological datasets Prediction of sample categories based on genomic profiles Research applications requiring robust classification of biological data Key Advantages: High-dimensional handling: SVMs excel with genomic data's typical high feature-to-sample ratios Integrated feature selection: Reduces noise and computational overhead while identifying key markers Probability estimation: Provides confidence scores essential for biological interpretation Validation robustness: Leave-one-out cross-validation ensures reliable performance metrics Operational flexibility: Multiple analysis modes support different research phases from exploration to prediction

Stecca Steindorff, Andrei [Lawrence Berkeley Natio↗
DOE OSTI2025

Reward Driven Workflows for Unsupervised Explainable Analysis of Phases and Ferroic Variants From Atomically Resolved Imaging Data

Rapid progress in aberration corrected electron microscopy necessitates development of robust methods for the identification of phases, ferroic variants, and other pertinent aspects of materials structure from imaging data. While unsupervised methods for clustering and classification are widely used for these tasks, their performance can be sensitive to hyperparameter selection in the analysis workflow. In this study, the effects of descriptors and hyperparameters are explored on the capability of unsupervised ML methods to distill local structural information, exemplified by the discovery of polarization and lattice distortion in Sm − dopped BiFeO 3 (BFO) thin films. It is demonstrated that a reward-driven approach can be used to optimize these key hyperparameters across the full workflow, where rewards are designed to reflect domain wall continuity and straightness, ensuring that the analysis aligns with the material's physical behavior. This approach allows the discovery of local descriptors that are best aligned with the specific physical behavior, providing insight into the fundamental physics of materials. The reward driven workflow is further extended to disentangle structural factors of variation via an optimized variational autoencoder (VAE). Lastly, the importance of well-defined rewards is explored as a quantifiable measure of the success of the workflow.

Barakati, Kamyar [University of Tennessee, Knoxvil↗
DOE OSTI2018

Topological Insulators versus Topological Dirac Semimetals in Honeycomb Compounds

Intriguing physical properties of materials stem from their chemical constituents, whereas the connection between them is often not clear. Here, we uncover a general chemical classification for the two quantum phases in the honeycomb ABX structure—topological insulator (TI) and topological Dirac semimetal (TDSM). First, we find among the 816 (existing as well as hypothetical) calculated compounds, 160 TIs (none were noted before), 96 TDSMs, 282 normal insulators (NIs), and 278 metals. Second, based on this classification, we have distilled a simple chemical regularity based on compound formulas for the selectivity between TI and TDSM: the ABX compounds that are TDSM have B atoms (part of the BX honeycomb layers) that come from the periodic table columns XI (Cu, Ag, Au) or XII (Zn, Cd, Hg), or Mg (group II), whereas the ABX compounds whose B at-oms come from columns I (Li, Na, K, Rb, Cs) or II (Ca, Sr, Ba) are TIs. Third, focusing on the ABX bismide compounds that are thermodynamically stable, we find a structural motif that delivers topological insulation and stability at the same time. Furthermore, this study opens the way to simultaneously design new topological materials based on the compositional rules indicated here.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Scenario Storyline Discovery for Planning in Multi‐Actor Human‐Natural Systems Confronting Change

Scenarios have emerged as valuable tools in managing complex human-natural systems, but the traditional approach of limiting focus on a small number of predetermined scenarios can inadvertently miss consequential dynamics, extremes, and diverse stakeholder impacts. Exploratory modeling approaches have been developed to address these issues by exploring a wide range of possible futures and identifying those that yield consequential vulnerabilities. However, vulnerabilities are typically identified based on aggregate robustness measures that do not take full advantage of the richness of the underlying dynamics in the large ensembles of model simulations and can make it hard to identify key dynamics and/or storylines that can guide planning or further analyses. This study introduces the FRamework for Narrative Storylines and Impact Classification (FRNSIC; pronounced “forensic”): a scenario discovery framework that addresses these challenges by organizing and investigating consequential scenarios using hierarchical classification of diverse outcomes across actors, sectors, and scales, while also aiding in the selection of scenario storylines, based on system dynamics that drive consequential outcomes. We present an application of this framework to the Upper Colorado River Basin, focusing on decadal droughts and their water scarcity implications for the basin's diverse users and its obligations to downstream states through Lake Powell. We show how FRNSIC can explore alternative sets of impact metrics and drought dynamics and use them to identify drought scenario storylines, that can be used to inform future adaptation planning.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2011

Formation and Evolution of the Disk System of the Milky Way: [alpha/Fe] Ratios and Kinematics of the SEGUE G-Dwarf Sample

We employ measurements of the [{alpha}/Fe] ratio fromlow-resolution (R {approx} 2000) spectra of 17,500 G-type dwarfs included in SDSS Data Release 8, selected using simple and well-understood selection criteria, to separate them into likely thin- and thick-disk subsamples. This classification, based on chemistry, is strongly motivated by the bi-modal distribution of stars in the [{alpha}/Fe] vs. [Fe/H] diagram. The resulting subsamples allow, for the first time, investigations of the kinematic behavior of thin- and thick-disk stars as a function of metallicity and position up to distances of 3 kpc from the Galactic plane. Both subsamples exhibit strong gradients of orbital rotational velocity with metallicity, but with opposite signs (-20 to -30 km s{sup -1} dex{sup -1} for the thin-disk population, and +40 to +50 km s{sup -1} dex{sup -1} for the thick-disk population). We find that the rotational velocity decreases with the distance from the plane for both disk components, with similar slopes (10 km s{sup -1} kpc{sup -1}), and a nearly constant difference in the mean rotational velocity of about 30 km s{sup -1}. The mean rotational velocity is uncorrelated with Galactocentric distance for the thin-disk subsample, and exhibits only a marginally significant correlation for the thick-disk subsample. Thick-disk stars exhibit a very strong trend of orbital eccentricity with metallicity (-0.2 dex{sup -1}), while the eccentricity does not change with metallicity for the thin-disk subsample. The eccentricity is almost independent of Galactocentric radius for the thin-disk stars, while a marginal gradient of the eccentricity with distance exists for the thick-disk population. Both subsamples possess similar trends of increasing eccentricity with distance from the Galactic plane, with a constant difference of about 0.1. The shapes of the overall distributions of orbital eccentricity for the thin- and thick-disk populations are quite different from one another, independent of distance from the plane; neither subsample has significant numbers of stars with eccentricity above 0.6. These observational results provide strong new constraints on models for the formation and evolution of the Milky Way's disk system. For example, the observed dependence of the mean rotational velocity on metallicity for thin-disk stars is inconsistent with predictions from classical local chemical evolution models. We also consider the predictions of several contemporary models of disk evolution, such as radial migration, gas-rich mergers, disk heating, and pure accretion models. We find that radial migration appears to have played an important role in the evolution of the thin-disk population, but possibly less so, relative to the gas-rich merger or disk heating scenarios, for the thick disk. Pure accretion models appear to be ruled out by the observed distribution of eccentricities for thick-disk stars. We emphasize that more physically realistic models, and simulations that probe a greater range of disk formation scenarios, need to be constructed in order to carry out the detailed quantitative comparisons that our new data enable.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2013

The role of galaxy interaction in the SFR-M {sub *} relation: characterizing morphological properties of Herschel-selected galaxies at 0.2 < z < 1.5

Galaxy interactions/mergers have been shown to dominate the population of IR-luminous galaxies (L {sub IR} ≳ 10{sup 11.6} L {sub ☉}) in the local universe (z ≲ 0.25). Recent studies based on the relation between galaxies' star formation rates and stellar mass (the SFR-M {sub *} relation or the {sup g}alaxy main sequence{sup )} have suggested that galaxy interaction/mergers may only become significant when galaxies fall well above the galaxy main sequence. Since the typical SFR at a given M {sub *} increases with redshift, the existence of the galaxy main sequence implies that massive, IR-luminous galaxies at high z may not necessarily be driven by galaxy interactions. We examine the role of galaxy interactions in the SFR-M {sub *} relation by carrying out a morphological analysis of 2084 Herschel-selected galaxies at 0.2 < z < 1.5 in the COSMOS field. Using a detailed visual classification scheme, we show that the fraction of 'disk galaxies' decreases and the fraction of 'irregular' galaxies increases systematically with increasing L {sub IR} out to z ≲ 1.5 and z ≲ 1.0, respectively. At L {sub IR} >10{sup 11.5} L {sub ☉}, ≳ 50% of the objects show evident features of strongly interacting/merger systems, where this percentage is similar to the studies of local IR-luminous galaxies. The fraction of interacting/merger systems also systematically increases with the deviation from the SFR-M {sub *} relation, supporting the view that galaxies falling above the main sequence are more dominated by mergers than the main-sequence galaxies. Meanwhile, we find that ≳ 18% of massive IR-luminous 'main-sequence galaxies' are classified as interacting systems, where this population may not evolve through the evolutionary track predicted by a simple gas exhaustion model.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Optimizing deep learning hyper-parameters through an evolutionary algorithm. In: MLHPC '15 Proceedings of the Workshop on Machine Learning in High-Performance Computing Environments, Article No. 4

There has been a recent surge of success in utilizing Deep Learning (DL) in imaging and speech applications for its relatively automatic feature generation and, in particular for convolutional neural networks (CNNs), high accuracy classification abilities. While these models learn their parameters through data-driven methods, model selection (as architecture construction) through hyper-parameter choices remains a tedious and highly intuition driven task. To address this, Multi-node Evolutionary Neural Networks for Deep Learning (MENNDL) is proposed as a method for automating network selection on computational clusters through hyper-parameter optimization performed via genetic algorithms.

Young, Steven R.↗
DOE OSTI2022

Quenching in the Right Place at the Right Time: Tracing the Shared History of Starbursts, Active Galactic Nuclei, and Poststarburst Galaxies Using Their Structures and Multiscale Environments

This work uses multiscale environments and the structures of galaxies in the Sloan Digital Sky Survey as consistency checks on the evolution from starburst to quiescence at redshift z < 0.2. The environmental indicators include fixed-aperture mass overdensities (δ xMpc , x ϵ {0.5, 1, 2, 4, 8} h –1 Mpc), k-nearest neighbor distances, the tidal parameter, halo mass (M h ), and satellite/central classification. The residuals of the specific star formation rates (Δ SSFRs) are used to select starbursts (ΔSSFR > 0.6 dex, N ≈ 8600). Quenched poststarbursts (QPSBs) are selected using Hα < 3 Å emission and Hδ A > 4 Å absorption (N ≈ 750). The environments of starbursts and QPSBs are compared with those of active galactic nuclei (AGNs) and inactive galaxies of varying ΔSSFR. The environments of starbursts, AGNs, and QPSBs are unlike the environments of most quiescent galaxies (QGs). About 70%–90% of starbursts, AGNs with Hδ A > 4 Å, and QPSBs are centrals; ~80%–90% have M h < 10 13 M ⊙ , and only ~2%–4% have M h > 10 14 M ⊙ or live in clusters. Their M h and satellite fractions are also different from those of QGs. All QPSBs are matched to some SFGs, starbursts, AGNs, and QGs of similar M $\star$ , environments, concentration indices, and velocity dispersions. A significant fraction (~20%–30%) of starbursts cannot be matched to QPSBs or QGs. The implications are that: (1) some starbursts do not quench rapidly; (2) the satellite-quenching mechanisms operating in high-density environments cannot account for most QPSBs; (3) the evolution from starbursts to QPSBs to QGs is not the dominant path at z < 0.2; and (4) starbursts are not mainly triggered by tidal interactions.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Public water supply infrastructure extensification and diversification in surface waters is insufficient to meet future demands in Texas

The data were developed to evaluate the capacity of existing and potential new surface water supply infrastructure to meet projected public water demands across districts in Texas under multiple future socioeconomic and climate scenarios. The database integrates hydrologic, water quality, infrastructure, energy, cost, demographic, and demand-projection information for candidate surface water supply locations. Candidate sites include stream reaches, waterbodies, reservoir surplus locations, and potential new reservoir sites. Water availability is characterized using historical and projected flow conditions, while site suitability is evaluated using five indicators: Water Availability Index (WAI), Water Quality Index (WQI), Energy Requirement Index (ERI), Water Treatment Cost (WTC), and Water Infrastructure Cost (WIC). The datasets include statewide candidate-site information, district-level demand projections under Shared Socioeconomic Pathways (SSPs), runoff-based allocation constraints, climate-stress metrics, and optimization outputs evaluating alternative infrastructure planning strategies. Optimization results compare Business-as-Usual (BAU) and All Surface Water (AllSW) demand-management approaches under both scaled and fixed cost-cap strategies. Associated validation datasets provide district-level feasibility assessments, infrastructure selection outcomes, cost-cap utilization, demand satisfaction metrics, and constraint diagnostics. Additional datasets quantify projected changes in storage and flow conditions as well as water availability stress for both existing and newly selected intake locations under the SSP5 scenario for mid-century and late-century climate conditions. Together, these datasets support assessment of the extent to which surface-water infrastructure expansion and diversification strategies can satisfy future public water demands while accounting for hydrologic, economic, and planning constraints across Texas. Dataset(s) Description Dataset_preoptimization.xlsx Comprehensive pre-optimization dataset containing candidate water-supply sites and associated hydrologic, water-quality, infrastructure, climate, demographic, runoff, and demand-projection variables used as inputs to the optimization analyses. Includes variable descriptions and the full statewide candidate-site database. District_level_site_selection.zip - Compressed archive containing all SSP-specific district-level optimization result files MESIO_ssp1_results.xlsx District-level site selection results for SSP1 (MESIO). Includes variable descriptions, BAU and AllSW site-selection results under scaled and fixed cost strategies, and district-level validation diagnostics. MESID_ssp2_results.xlsx District-level site selection results for SSP2 (MESID). Includes variable descriptions, BAU and AllSW site-selection results under scaled and fixed cost strategies, and district-level validation diagnostics. LCMRD_ssp3_results.xlsx District-level site selection results for SSP3 (LCMRD). Includes variable descriptions, BAU and AllSW site-selection results under scaled and fixed cost strategies, and district-level validation diagnostics. IRDev-Low_ssp4l_results.xlsx District-level site selection results for SSP4-Low (IRDev-Low). Includes variable descriptions, BAU and AllSW site-selection results under scaled and fixed cost strategies, and district-level validation diagnostics. IRDev-High_ssp4h_results.xlsx District-level site selection results for SSP4-High (IRDev-High). Includes variable descriptions, BAU and AllSW site-selection results under scaled and fixed cost strategies, and district-level validation diagnostics. RSIM_ssp5_results.xlsx District-level site selection results for SSP5 (RSIM). Includes variable descriptions, BAU and AllSW site-selection results under scaled and fixed cost strategies, and district-level validation diagnostics. tx_hydrological_stress.xlsx Hydrological stress dataset for existing and newly selected intake locations. Includes projected mid-century and late-century changes, gain/loss classifications, planning strategy information, and accompanying variable descriptions. Also includes water-stress metrics derived from historical and projected low-flow conditions.

Okoye, Perpetua I. (ORCID:0000000215545033)↗
DOE OSTI2023

Batch Active Learning for Multispectral and Hyperspectral Image Segmentation Using Similarity Graphs

Abstract Graph learning, when used as a semi-supervised learning (SSL) method, performs well for classification tasks with a low label rate. We provide a graph-based batch active learning pipeline for pixel/patch neighborhood multi- or hyperspectral image segmentation. Our batch active learning approach selects a collection of unlabeled pixels that satisfy a graph local maximum constraint for the active learning acquisition function that determines the relative importance of each pixel to the classification. This work builds on recent advances in the design of novel active learning acquisition functions (e.g., the Model Change approach in arXiv:2110.07739) while adding important further developments including patch-neighborhood image analysis and batch active learning methods to further increase the accuracy and greatly increase the computational efficiency of these methods. In addition to improvements in the accuracy, our approach can greatly reduce the number of labeled pixels needed to achieve the same level of the accuracy based on randomly selected labeled pixels.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2010

A signature dissimilarity measure for trabecular bone texture in knee radiographs

Purpose: The purpose of this study is to develop a dissimilarity measure for the classification of trabecular bone (TB) texture in knee radiographs. Problems associated with the traditional extraction and selection of texture features and with the invariance to imaging conditions such as image size, anisotropy, noise, blur, exposure, magnification, and projection angle were addressed. Methods: In the method developed, called a signature dissimilarity measure (SDM), a sum of earth mover's distances calculated for roughness and orientation signatures is used to quantify dissimilarities between textures. Scale-space theory was used to ensure scale and rotation invariance. The effects of image size, anisotropy, noise, and blur on the SDM developed were studied using computer generated fractal texture images. The invariance of the measure to image exposure, magnification, and projection angle was studied using x-ray images of human tibia head. For the studies, Mann-Whitney tests with significance level of 0.01 were used. A comparison study between the performances of a SDM based classification system and other two systems in the classification of Brodatz textures and the detection of knee osteoarthritis (OA) were conducted. The other systems are based on weighted neighbor distance using compound hierarchy of algorithms representing morphology (WND-CHARM) and local binary patterns (LBP). Results: Results obtained indicate that the SDM developed is invariant to image exposure (2.5-30 mA s), magnification (x1.00-x1.35), noise associated with film graininess and quantum mottle (<25%), blur generated by a sharp film screen, and image size (>64x64 pixels). However, the measure is sensitive to changes in projection angle (>5 deg.), image anisotropy (>30 deg.), and blur generated by a regular film screen. For the classification of Brodatz textures, the SDM based system produced comparable results to the LBP system. For the detection of knee OA, the SDM based system achieved 78.8% classification accuracy and outperformed the WND-CHARM system (64.2%). Conclusions: The SDM is well suited for the classification of TB texture images in knee OA detection and may be useful for the texture classification of medical images in general.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2025

Using active learning to improve quasar identification for the DESI spectra processing pipeline

The Dark Energy Spectroscopic Instrument (DESI) survey uses an automatic spectral classification pipeline to classify spectra. QuasarNET is a convolutional neural network used as part of this pipeline originally trained using data from the Baryon Oscillation Spectroscopic Survey (BOSS). In this paper we implement an active learning algorithm to optimally select spectra to use for training a new version of the QuasarNET weights file using only DESI data, with the goal of improving classification accuracy. This active learning algorithm includes a novel outlier rejection step using a Self-Organizing Map to ensure we label spectra representative of the larger quasar sample observed in DESI. We perform two iterations of the active learning pipeline, assembling a final dataset of 5600 labeled spectra, a small subset of the approximately 1.3 million quasar targets in DESI's Data Release 1. When splitting the spectra into training and validation subsets we achieve similar performance to the previously trained weights file in completeness and purity calculated on the validation dataset but do so with less than one tenth of the amount of training data. The new weights also more consistently classify objects in the same way when used on unlabeled data compared to the old weights file. In the process of improving QuasarNET's classification accuracy we discovered a systemic error in QuasarNET's redshift estimation and used our findings to improve our understanding of QuasarNET's redshifts.

Machine learning↗
DOE OSTI2010

Results from the Supernova Photometric Classification Challenge

We report results from the Supernova Photometric Classification Challenge (SNPCC), a publicly released mix of simulated supernovae (SNe), with types (Ia, Ibc, and II) selected in proportion to their expected rate. The simulation was realized in the griz filters of the Dark Energy Survey (DES) with realistic observing conditions (sky noise, point-spread function and atmospheric transparency) based on years of recorded conditions at the DES site. Simulations of non-Ia type SNe are based on spectroscopically confirmed light curves that include unpublished non-Ia samples donated from the Carnegie Supernova Project (CSP), the Supernova Legacy Survey (SNLS), and the Sloan Digital Sky Survey-II (SDSS-II). A spectroscopically confirmed subset was provided for training. We challenged scientists to run their classification algorithms and report a type and photo-z for each SN. Participants from 10 groups contributed 13 entries for the sample that included a host-galaxy photo-z for each SN, and 9 entries for the sample that had no redshift information. Several different classification strategies resulted in similar performance, and for all entries the performance was significantly better for the training subset than for the unconfirmed sample. For the spectroscopically unconfirmed subset, the entry with the highest average figure of merit for classifying SNe~Ia has an efficiency of 0.96 and an SN~Ia purity of 0.79. As a public resource for the future development of photometric SN classification and photo-z estimators, we have released updated simulations with improvements based on our experience from the SNPCC, added samples corresponding to the Large Synoptic Survey Telescope (LSST) and the SDSS, and provided the answer keys so that developers can evaluate their own analysis.

79 ASTRONOMY AND ASTROPHYSICS↗