SEARCH · Engineering Papers
Results for “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.
CONSENSUS INFERENCE ON MOBILE PHONE SENSORS FOR ACTIVITY RECOGNITION
Abstract not provided
Sub-class Recognition from Aggregate Class Labels: Preliminary Results
In many practical situations it is not feasible to collect labeled samples for all available classes in a domain. Especially in supervised classification of remotely sensed images it is impossible to collect ground truth information over large geographic regions for all thematic classes. As a result often analysts collect labels for aggregate classes. In this paper we present a novel learning scheme that automatically learns sub-classes from the user given aggregate classes. We model each aggregate class as finite Gaussian mixture instead of classical assumption of unimodal Gaussian per class. The number of components in each finite Gaussian mixture are automatically estimated. Experimental results on real remotely sensed image classification showed not only improved accuracy in aggregate class classification but the proposed method also recognized sub-classes.
From Data to Insights: A Covariate Analysis of the IARPA BRIAR Dataset for Multimodal Biometric Recognition Algorithms at Altitude and Range
This paper examines covariate effects on fused whole body biometrics performance in the IARPA BRIAR dataset, specifically focusing on UAV platforms, elevated positions, and distances up to 1000 meters. The dataset includes outdoor videos compared with indoor images and controlled gait recordings. Normalized raw fusion scores relate directly to predicted false accept rates (FAR), offering an intuitive means for interpreting model results. A linear model is developed to predict biometric algorithm scores, analyzing their performance to identify the most influential covariates on accuracy at altitude and range. Weather factors like temperature, wind speed, solar loading, and turbulence are also investigated in this analysis. The study found that resolution and camera distance best predicted accuracy and findings can guide future research and development efforts in long-range/elevated/UAV biometrics and support the creation of more reliable and robust systems for national security and other critical domains.
Remote Recognition of In-Bed Postures Using a Thermopile Array Sensor With Machine Learning
Not provided.
Robust Foot Motion Recognition Using Stride Detection and Weak Supervision-Based Fast Labeling
Not provided.
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.
LINEBACKER: LINE-speed Bio-inspired Analysis and Characterization for Event Recognition
The cyber world is a complex domain, with digital systems mediating a wide spectrum of human and machine behaviors. While this is enabling a revolution in the way humans interact with each other and data, it also is exposing previously unreachable infrastructure to a worldwide set of actors. Existing solutions for intrusion detection and prevention that are signature-focused typically seek to detect anomalous and/or malicious activity for the sake of preventing or mitigating negative impacts. But a growing interest in behavior-based detection is driving new forms of analysis that move the emphasis from static indicators (e.g. rule-based alarms or tripwires) to behavioral indicators that accommodate a wider contextual perspective. Similar to cyber systems, biosystems have always existed in resource-constrained hostile environments where behaviors are tuned by context. So we look to biosystems as an inspiration for addressing behavior-based cyber challenges. In this paper, we introduce LINEBACKER, a behavior-model based approach to recognizing anomalous events in network traffic and present the design of this approach of bio-inspired and statistical models working in tandem to produce individualized alerting for a collection of systems. Preliminary results of these models operating on historic data are presented along with a plugin to support real-world cyber operations.
Remote Sensor Design for Visual Recognition With Convolutional Neural Networks
Not Available
Video Action Recognition with an Additional End-To-End Trained Temporal Stream
Explore the source record for details and available documents.
Recognition of dominant attractants by key chemoreceptors mediates recruitment of plant growth‐promoting rhizobacteria
Summary Chemotaxis to plant root exudates is supposed to be a prerequisite for efficient root colonization by rhizobacteria. This is a highly multifactorial process since root exudates are complex compound mixtures of which components are recognized by different chemoreceptors. Little information is available as to the key components in root exudates and their receptors that drive colonization related chemotaxis. We present here the first global assessment of this issue using the plant growth‐promoting rhizobacterium (PGPR) Bacillus velezensis SQR9 (formerly B. amyloliquefaciens ). This strain efficiently colonizes cucumber roots, and here, we show that chemotaxis to cucumber root exudates was essential in this process. We conducted chemotaxis assays using cucumber root exudates at different concentrations, individual exudate components as well as recomposed exudates, taking into account their concentrations detected in root exudates. Results indicated that two key chemoreceptors, McpA and McpC, were essential for root exudate chemotaxis and root colonization. Both receptors possess a broad ligand range and recognize most of the exudate key components identified (malic, fumaric, gluconic and glyceric acids, Lys, Ser, Ala and mannose). The remaining six chemoreceptors did not contribute to exudate chemotaxis. This study provides novel insight into the evolution of the chemotaxis system in rhizobacteria.
Understanding molecular recognition of promiscuity of thermophilic methionine adenosyltransferase sMAT from Sulfolobus solfataricus
Abstract not provided
Structural basis for substrate recognition and inhibition of prephenate aminotransferase from Arabidopsis
Abstract not provided
Recognition of Cognate Transfer RNA by the 30S Ribosomal Subunit
Crystal structures of the 30S ribosomal subunit in complex with messenger RNA and cognate transfer RNA in the A site, both in the presence and absence of the antibiotic paromomycin, have been solved at between 3.1 and 3.3 angstroms resolution. Cognate transfer RNA (tRNA) binding induces global domain movements of the 30S subunit and changes in the conformation of the universally conserved and essential bases A1492, A1493, and G530 of 16S RNA. These bases interact intimately with the minor groove of the first two base pairs between the codon and anticodon, thus sensing Watson-Crick base-pairing geometry and discriminating against near-cognate tRNA. The third, or 'wobble,' position of the codon is free to accommodate certain noncanonical base pairs. By partially inducing these structural changes, paromomycin facilitates binding of near-cognate tRNAs.
Structure of Tracheal Cytotoxin in Complex with a Heterodimeric Pattern-Recognition Receptor
Abstract not provided
Antibody Recognition of a Highly Conserved Influenza Virus Epitope
Influenza virus presents an important and persistent threat to public health worldwide, and current vaccines provide immunity to viral isolates similar to the vaccine strain. High-affinity antibodies against a conserved epitope could provide immunity to the diverse influenza subtypes and protection against future pandemic viruses. Cocrystal structures were determined at 2.2 and 2.7 angstrom resolutions for broadly neutralizing human antibody CR6261 Fab in complexes with the major surface antigen (hemagglutinin, HA) from viruses responsible for the 1918 H1N1 influenza pandemic and a recent lethal case of H5N1 avian influenza. In contrast to other structurally characterized influenza antibodies, CR6261 recognizes a highly conserved helical region in the membrane-proximal stem of HA1 and HA2. The antibody neutralizes the virus by blocking conformational rearrangements associated with membrane fusion. The CR6261 epitope identified here should accelerate the design and implementation of improved vaccines that can elicit CR6261-like antibodies, as well as antibody-based therapies for the treatment of influenza.