Search NASA⌕ Search

SEARCH · Search NASA

Results for “normalized mutual information”

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.

11 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2016

Cross Correlation versus Normalized Mutual Information on Image Registration

This is the first study to quantitatively assess and compare cross correlation and normalized mutual information methods used to register images in subpixel scale. The study shows that the normalized mutual information method is less sensitive to unaligned edges due to the spectral response differences than is cross correlation. This characteristic makes the normalized image resolution a better candidate for band to band registration. Improved band-to-band registration in the data from satellite-borne instruments will result in improved retrievals of key science measurements such as cloud properties, vegetation, snow and fire.

band-to-band↗
DOE OSTI2023

Improved quality metrics for association and reproducibility in chromatin accessibility data using mutual information

Correlation metrics are widely utilized in genomics analysis and often implemented with little regard to assumptions of normality, homoscedasticity, and independence of values. This is especially true when comparing values between replicated sequencing experiments that probe chromatin accessibility, such as assays for transposase-accessible chromatin via sequencing (ATAC-seq). Such data can possess several regions across the human genome with little to no sequencing depth and are thus non-normal with a large portion of zero values. Despite distributed use in the epigenomics field, few studies have evaluated and benchmarked how correlation and association statistics behave across ATAC-seq experiments with known differences or the effects of removing specific outliers from the data. Here, we developed a computational simulation of ATAC-seq data to elucidate the behavior of correlation statistics and to compare their accuracy under set conditions of reproducibility. Using these simulations, we monitored the behavior of several correlation statistics, including the Pearson’s R and Spearman’s ρ coefficients as well as Kendall’s τ and Top–Down correlation. We also test the behavior of association measures, including the coefficient of determination R 2 , Kendall’s W, and normalized mutual information. Our experiments reveal an insensitivity of most statistics, including Spear man’s ρ, Kendall’s τ, and Kendall’s W, to increasing differences between simulated ATAC-seq replicates. The removal of co-zeros (regions lacking mapped sequenced reads) between simulated experiments greatly improves the estimates of correlation and association. After removing co-zeros, the R 2 coefficient and normalized mutual information display the best performance, having a closer one-to-one relationship with the known portion of shared, enhanced loci between simulated replicates. When comparing values between experimental ATAC-seq data using a random forest model, mutual information best predicts ATAC-seq replicate relationships. Collectively, this study demonstrates how measures of correlation and association can behave in epigenomics experiments. We provide improved strategies for quantifying relationships in these increasingly prevalent and important chromatin accessibility assays.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRSDate not supplied

Automated Registration of Multi-Mode Nondestructive Evaluation Data

Registration techniques play a central role in applications of image processing to computer vision, medical imaging, and automatic target tracking. Feature-based techniques such as scale-invariant feature transform (SIFT) and speeded up robust features (SURF) are commonly used to register images derived from a single modality. However, SIFT and SURF struggle to register images from different modalities because the features tend to manifest rather differently and at sometimes very different length-scales. The most successful methods that have been developed to register multi-modal data use information-theoretic approaches. These methods play a key part in nondestructive evaluation scenarios where data that is collected by sensors of different modalities must be registered to be fused. In this paper, automated registration based on normalized mutual information is applied to align data derived from ultrasonic and radiographic inspections of (i) additively manufactured titanium alloy test coupons, and (ii) thin, lithium metal pouch-cell batteries. The quality of the registration is quantified in terms of computational resources and spatial accuracy. In the first case the X-ray computed tomography (XCT) data is captured on a region corresponding to a small subset of the ultrasonic data, while in the case of the lithium batteries the digital radiography (DR) captures a larger region of interest than the ultrasonic data. In both cases the radiographic data resolution is much higher than for ultrasound, but interestingly, in both cases the accuracy of the registration is approximately equal to two-to-three-pixel lengths in the ultrasonic images.

Nondestructive Evaluation↗
DOE OSTI2024

Transient anisotropic kernel for probabilistic learning on manifolds

PLoM (Probabilistic Learning on Manifolds) is a method introduced in 2016 for handling small training datasets by projecting an Itô equation from a stochastic dissipative Hamiltonian dynamical system, acting as the MCMC generator, for which the KDE-estimated probability measure with the training dataset is the invariant measure. PLoM performs a projection on a reduced-order vector basis related to the training dataset, using the diffusion maps (DMAPS) basis constructed with a time-independent isotropic kernel. In this paper, we propose a new ISDE projection vector basis built from a transient anisotropic kernel, providing an alternative to the DMAPS basis to improve statistical surrogates for stochastic manifolds with heterogeneous data. The construction ensures that for times near the initial time, the DMAPS basis coincides with the transient basis. For larger times, the differences between the two bases are characterized by the angle of their spanned vector subspaces. The optimal instant yielding the optimal transient basis is determined using an estimation of mutual information from Information Theory, which is normalized by the entropy estimation to account for the effects of the number of realizations used in the estimations. Consequently, this new vector basis better represents statistical dependencies in the learned probability measure for any dimension. Three applications with varying levels of statistical complexity and data heterogeneity validate the proposed theory, showing that the transient anisotropic kernel improves the learned probability measure.

Diffusion maps↗
DOE OSTI2026

Dark Energy Survey Year 3 results: Simulation-based 𝑤CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Data-driven approaches using deep learning are emerging as powerful techniques to extract non-Gaussian information from cosmological large-scale structure. Here, this work presents the first simulation-based inference (SBI) pipeline that combines weak lensing and galaxy clustering maps in a realistic Dark Energy Survey Year 3 (DES Y3) configuration and serves as preparation for a forthcoming analysis of the survey data. We develop a scalable forward model based on the CosmoGridV1 suite of N-body simulations to generate over one million self-consistent mock realizations of DES Y3 at the map level. Leveraging this large dataset, we train deep graph convolutional neural networks on the full survey footprint in spherical geometry to learn low-dimensional features that approximately maximize mutual information with target parameters. These learned compressions enable neural density estimation of the implicit likelihood via normalizing flows in a ten-dimensional parameter space spanning cosmological 𝑤CDM, intrinsic alignment, and linear galaxy bias parameters, while marginalizing over baryonic, photometric redshift, and shear bias nuisances. To ensure robustness, we extensively validate our inference pipeline using synthetic observations derived from both systematic contaminations in our forward model and independent Buzzard galaxy catalogs. Our forecasts yield significant improvements in cosmological parameter constraints, achieving 2−3× higher figures of merit in the 𝛺 𝑚 − 𝑆 8 plane relative to our implementation of baseline two-point statistics and effectively breaking parameter degeneracies through probe combination. These results demonstrate the potential of SBI analyses powered by deep learning for upcoming Stage-IV wide-field imaging surveys.

Thomsen, A. [Zurich, ETH] (ORCID:0000000203099021)↗
NASA NTRS1981

Maximum likelihood clustering with dependent feature trees

The decomposition of mixture density of the data into its normal component densities is considered. The densities are approximated with first order dependent feature trees using criteria of mutual information and distance measures. Expressions are presented for the criteria when the densities are Gaussian. By defining different typs of nodes in a general dependent feature tree, maximum likelihood equations are developed for the estimation of parameters using fixed point iterations. The field structure of the data is also taken into account in developing maximum likelihood equations. Experimental results from the processing of remotely sensed multispectral scanner imagery data are included.

Chittineni, C. B.↗
NASA NTRS2014

Revised (Mixed-Effects) Estimation for Forest Burning Emissions of Gases and Smoke, Fire/Emission Factor Typology, and Potential Remote Sensing Classification of Types for Ozone and Black-Carbon Simulation

We summarize recent progress (a) in correcting biomass burning emissions factors deduced from airborne sampling of forest fire plumes, (b) in understanding the variability in reactivity of the fresh plumes sampled in ARCTAS (2008), DC3 (2012), and SEAC4RS (2013) airborne missions, and (c) in a consequent search for remotely sensed quantities that help classify forest-fire plumes. Particle properties, chemical speciation, and smoke radiative properties are related and mutually informative, as pictures below suggest (slopes of lines of same color are similar). (a) Mixed-effects (random-effects) statistical modeling provides estimates of both emission factors and a reasonable description of carbon-burned simultaneously. Different fire plumes will have very different contributions to volatile organic carbon reactivity; this may help explain differences of free NOx(both gas- and particle-phase), and also of ozone production, that have been noted for forest-fire plumes in California. Our evaluations check or correct emission factors based on sequential measurements (e.g., the Normalized Ratio Enhancement and similar methods). We stress the dangers of methods relying on emission-ratios to CO. (b) This work confirms and extends many reports of great situational variability in emissions factors. VOCs vary in OH reactivity and NOx-binding. Reasons for variability are not only fuel composition, fuel condition, etc., but are confused somewhat by rapid transformation and mixing of emissions. We use "unmixing" (distinct from mixed-effects) statistics and compare briefly to approaches like neural nets. We focus on one particularly intense fire the notorious Yosemite Rim Fire of 2013. In some samples, NOx activity was not so suppressed by binding into nitrates as in other fires. While our fire-typing is evolving and subject to debate, the carbon-burned delta(CO2+CO) estimates that arise from mixed effects models, free of confusion by background-CO2 variation, should provide a solid base for discussion. (c) We report progress using promising links we find between emissions-related "fire types" and promising features deducible from remote observations of plumes, e.g., single scatter albedo, Angstrom exponent of scattering, Angstrom exponent of absorption, (CO column density)/(aerosol optical depth).

Remote sensing↗
NASA NTRS1988

Mass transport phenomena between bubbles and dissolved gases in liquids under reduced gravity conditions

This paper will describe the experimental and analytical work that has been done to establish justification and feasibility for a Shuttle mid-deck experiment involving mass transfer between a gas bubble and a liquid. The experiment involves the observation and measurement of the dissolution of an isolated, immobile gas bubble of specified size and composition in a thermostatted solvent liquid of known concentration in the reduced gravity environment of earth orbit. Methods to generate and deploy the bubble have been successful both in normal gravity using mutually buoyant fluids and under reduced gravity conditions in the NASA Lear Jet. Initialization of the experiment with a bubble of a prescribed size and composition in a liquid of known concentration has been accomplished using the concept of unstable equilibrium. Subsequent bubble dissolution or growth is obtained by a step increase or decrease in the liquid pressure. A numerical model has been developed which simulates the bubble dynamics and can be used to determine molecular parameters by comparison with the experimental data. The primary objective of the experiment is the elimination of convective effects that occur in normal gravity. The results will yield information on transport under conditions of pure diffusion.

Dewitt, Kenneth J.↗
NASA NTRS2004

Exchange of Standardized Flight Dynamics Data

Spacecraft operations require the knowledge of the vehicle trajectory and attitude and also that of other spacecraft or natural bodies. This knowledge is normally provided by the Flight Dynamics teams of the different space organizations and, as very often spacecraft operations involve more than one organization, this information needs to be exchanged between Agencies. This is why the Navigation Working Group within the CCSDS (Consultative Committee for Space Data Systems), has been instituted with the task of establishing standards for the exchange of Flight Dynamics data. This exchange encompasses trajectory data, attitude data, and tracking data. The Navigation Working Group includes regular members and observers representing the participating Space Agencies. Currently the group includes representatives from CNES, DLR, ESA, NASA and JAXA. This Working Group meets twice per year in order to devise standardized language, methods, and formats for the description and exchange of Navigation data. Early versions of some of these standards have been used to support mutual tracking of ESA and NASA interplanetary spacecraft, especially during the arrival of the 2003 missions to Mars. This paper provides a summary of the activities carried out by the group, briefly outlines the current and envisioned standards, describes the tests and operational activities that have been performed using the standards, and lists and discusses the lessons learned from these activities.

standards↗
DOE OSTI2024

Informative and non-informative decomposition of turbulent flow fields

Not all the information in a turbulent field is relevant for understanding particular regions or variables in the flow. Here, we present a method for decomposing a source field into its informative Φ I (x, t) and residual Φ R (x, t) components relative to another target field. The method is referred to as informative and non-informative decomposition (IND). All the necessary information for physical understanding, reduced-order modelling and control of the target variable is contained in Φ I (x, t), whereas Φ R (x, t) offers no substantial utility in these contexts. The decomposition is formulated as an optimisation problem that seeks to maximise the time-lagged mutual information of the informative component with the target variable while minimising the mutual information with the residual component. The method is applied to extract the informative and residual components of the velocity field in a turbulent channel flow, using the wall shear stress as the target variable. We demonstrate the utility of IND in three scenarios: (i) physical insight into the effect of the velocity fluctuations on the wall shear stress; (ii) prediction of the wall shear stress using velocities far from the wall; and (iii) development of control strategies for drag reduction in a turbulent channel flow using opposition control. In case (i), IND reveals that the informative velocity related to wall shear stress consists of wall-attached high- and low-velocity streaks, collocated with regions of vertical motions and weak spanwise velocity. This informative structure is embedded within a larger-scale streak–roll structure of residual velocity, which bears no information about the wall shear stress. In case (ii), the best-performing model for predicting wall shear stress is a convolutional neural network that uses the informative component of the velocity as input, while the residual velocity component provides no predictive capabilities. Finally, in case (iii), we demonstrate that the informative component of the wall-normal velocity is closely linked to the observability of the target variable and holds the essential information needed to develop successful control strategies.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Electron-mediated entanglement of two distant macroscopic ferromagnets within a nonequilibrium spintronic device

Using the nascent concept of quantum spin-transfer torque, we demonstrate that a current pulse can be harnessed to entangle quantum localized spins of two spatially separated ferromagnets (FMs) which are initially unentangled. The envisaged setup is composed of a spin-polarizer (FM p ) and a spin-analyzer (FM a ) FM layers separated by a normal metal (NM) spacer. The injection of a current pulse into the device leads to a time-dependent superposition of many-body states characterized by a high degree of entanglement between the spin degrees of freedom of the two distant FM layers. The nonequilibrium dynamics are due to the transfer of spin angular momentum from itinerant electrons to the localized spins via a quantum spin-torque mechanism that remains active even for collinear but antiparallel arrangements of the FM p and FM a magnetizations (a situation in which the conventional spin torque is absent). We quantify the mixed-state entanglement generated between the FM layers by tracking the time evolution of the full density matrix and analyzing the build-up of the mutual logarithmic negativity over time. The effect of decoherence and dissipation in the FM layers due to coupling to bosonic baths at finite temperature, the use of multielectron current pulses, and the dependence on the number of spins are also considered in an effort to ascertain the robustness of our predictions under realistic conditions. Lastly, we propose a “current-pump–x-ray-probe” scheme, utilizing ultrafast x-ray spectroscopy, that can witness nonequilibrium and transient entanglement of the FM layers by extracting its time-dependent quantum Fisher information.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗