SEARCH · Engineering Papers

Results for “Labeling”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

All sources NASA NTRS DOE OSTI

NASA NTRS1980

Probabilistic cluster labeling of imagery data

The problem of obtaining the probabilities of class labels for the clusters using spectral and spatial information from a given set of labeled patterns and their neighbors is considered. A relationship is developed between class and clusters conditional densities in terms of probabilities of class labels for the clusters. Expressions are presented for updating the a posteriori probabilities of the classes of a pixel using information from its local neighborhood. Fixed-point iteration schemes are developed for obtaining the optimal probabilities of class labels for the clusters. These schemes utilize spatial information and also the probabilities of label imperfections. Experimental results from the processing of remotely sensed multispectral scanner imagery data are presented.

Chittineni, C. B.↗

NASA NTRS1991

The Doubly Labeled Water Method for Measuring Human Energy Expenditure: Adaptations for Spaceflight

It is essential to determine human energy requirements in space, and the doubly labeled water method has been identified as the most appropriate means of indirect calorimetry to meet this need. The method employs naturally occurring, stable isotopes of hydrogen (H-2, deuterium) and oxygen (O-18) which, after dosing, mix with body water. The deuterium is lost from the body as water while the O-18 is eliminated as both water and CO2. The difference between the two isotope elimination rates is therefore a measure of CO2 production and hence energy expenditure. Spaceflight will present a unique challenge to the application of the doubly labeled water method. Specifically, interpretation of doubly labeled water results assumes that the natural abundance or 'background' levels of the isotopes remain constant during the measurement interval. To address this issue, an equilibration model will be developed in an ongoing ground-based study. As energy requirements of women matched to counterparts in the Astronauts Corps are being determined by doubly labeled water, the baseline isotope concentration will be changed by consumption of 'simulated Shuttle water' which is artificially enriched. One group of subjects will be equilibrated on simulated Shuttle water prior to energy determinations by doubly labeled water while the others will consume simulated Shuttle water after dosing. This process will allow us to derive a prediction equation to mathematically model the effect of changing background isotope concentrations.

Schulz, Leslie O.↗

NASA NTRS1978

Secondary error analysis: The evaluation of analyst dot labeling

The author has identified the following significant results. From this examination of 25 test segments using Al labeling and ground truth labeling, the PCC on type 1 dots was found to be signficantly better for both types of ground truth labeled procedures than the PCC obtained using Al labeling. No significant difference in the PCC was found for type 2 dots. However, in all three treatments, the type 2 dots included pixels which fell on boundaries or were mixed pixels. This accounted for all PCC2 values being equally low. The proportion estimates achieved in these classifications showed no significant differences between procedures.

Havens, K. A.↗

NASA NTRS1979

Maximum likelihood estimation of label imperfections and its use in the identification of mislabeled patterns

The problem of estimating label imperfections and the use of the estimation in identifying mislabeled patterns is presented. Expressions for the maximum likelihood estimates of classification errors and a priori probabilities are derived from the classification of a set of labeled patterns. Expressions also are given for the asymptotic variances of probability of correct classification and proportions. Simple models are developed for imperfections in the labels and for classification errors and are used in the formulation of a maximum likelihood estimation scheme. Schemes are presented for the identification of mislabeled patterns in terms of threshold on the discriminant functions for both two-class and multiclass cases. Expressions are derived for the probability that the imperfect label identification scheme will result in a wrong decision and are used in computing thresholds. The results of practical applications of these techniques in the processing of remotely sensed multispectral data are presented.

Chittineni, C. B.↗

NASA NTRS1979

A programmed labeling approach to image interpretation

Manual labeling techniques require the analyst-interpreter to use not only production film converter products but also agricultural and meteorological data and spectral aids in an integrated, judgmental fashion. To control an anticipated high variance in these techniques, a semiautomatic labeling technology was developed. The product of this technology is label identification from statistical tabulation (LIST) which operates from a discriminant basis and has the ability to measure the reliability of the label and to introduce an arbitrary bias. The development of LIST and its properties are described. Numerical results of an application are included and the evaluation of LIST is discussed.

Pore, M. D.↗

NASA NTRS1981

Pixel labeling by supervised probabilistic relaxation

A simple modification to existing probabilistic relaxation procedures is suggested which allows the information contained in initial labels to exert an influence on the direction of relaxation throughout the process. In this manner, the initial labels assume more importance than with conventional algorithms and are used in combination with the outcome of relaxation at each iteration to produce a cooperative estimate of the correct label for a particular object. Pixel labeling examples are presented which show the performance that can be obtained with the modified algorithm. The procedure is readily generalized to allow other data to influence the process.

Richards, J. A.↗

NASA NTRS1988

Bar Code Labels

American Bar Codes, Inc. developed special bar code labels for inventory control of space shuttle parts and other space system components. ABC labels are made in a company-developed anodizing aluminum process and consecutively marketed with bar code symbology and human readable numbers. They offer extreme abrasion resistance and indefinite resistance to ultraviolet radiation, capable of withstanding 700 degree temperatures without deterioration and up to 1400 degrees with special designs. They offer high resistance to salt spray, cleaning fluids and mild acids. ABC is now producing these bar code labels commercially or industrial customers who also need labels to resist harsh environments.

Source record↗

NASA NTRS2019

Image Labeler: A Web Interface to Catalog Earth Science Events

Advances in machine learning (ML) have made it possible to automatically detect Earth science phenomena from satellite imagery. While useful, ML algorithms typically require an extensive dataset containing labeled images for training. Systematic labeling and management of such datasets is quite cumbersome. With this in mind, we present the Image Labeler. Image Labeler is a fast and scalable cloud-based tool that facilitates the rapid development of Earth science event databases, in order to aid automated ML-based image classification.

Case Study↗

NASA NTRSDate not supplied

ImageLabler: Labeling and Managing Image Data for Machine Learning in the Earth Sciences

While machine learning techniques for image classification have been around for a long time, storing and managing the vast number of images required as training data is still a problem for scientists. This is especially true for the field of Earth science, where only recently have experts begun using machine learning techniques for image-based phenomena classification. Image Labeler, a fast and scalable cloud-based tagging platform for Earth science images, seeks to improve upon existing methods of managing images and associated metadata, such as maintaining categorized folders of images on a local machine, a process that can be cumbersome and difficult to scale. The platform facilitates rapid development of image-based Earth science phenomena training datasets by allowing scientists to upload their existing imagery as well as extract new samples from open satellite imagery services made available through NASA’s Global Imagery Browse Service (GIBS). Image Labeler also supports GeoTIFF data, with capabilities such as displaying GeoTIFFs on an interactive map, drawing shapefiles over them, and tagging them with additional metadata. This allows scientists to perform spatiotemporal subsetting with geographic information and develop training data more quickly. Built using modern web technologies, Image Labeler includes additional capabilities such as team collaboration for large-scale image tagging projects. Users can download their data in a machine-learning-ready format, allowing scientists to spend time on experimentation rather than on the collection of training data. In this presentation, we demonstrate how Image Labeler seeks to become a one-stop image data management solution for machine learning applications in Earth science.

Ashish Acharya↗

NASA NTRS1976

Viking labeled release biology experiment - Interim results

All results of the labeled-release life-detection experiment conducted on Mars prior to conjunction are summarized. Tests at both landing sites provide remarkably similar evolution of radioactive gas upon addition of a radioactive nutrient to the Mars sample. The 'active' agent in the sample is stable to 18 C, but is substantially inactivated by heat treatment for 3 hours at 50 C and completely inactivated at 160 C, as would be anticipated if the active response were caused by microorganisms. Results from test and heat-sterilized control samples are compared with those obtained from terrestrial soils and a lunar sample. Possible nonbiological explanations of the Mars data are reviewed. Although such explanations of the labeled-release data depend on UV irradiation, the labeled-release response does not appear to depend on recent direct UV activation of surface material. Available facts do not yet permit a conclusion regarding the existence of life on Mars.

Levin, G. V.↗

NASA NTRS1978

Design of polymeric immunomicrospheres for cell labelling and cell separation

Synthesis of several classes of hydrophylic microspheres applied to cell labeling and cell separation is described. Five classes of cross-linked microspheres with functional groups such as carboxyl, hydroxyl, amide and/or pyridine groups were synthesized. These functional groups were used to bind covalently antibodies and other proteins to the surface of the microspheres. To optimize the derivatisation technique, polyglutaraldehyde immunomicrospheres were prepared and utilized. Specific populations of human and murine lymphocytes were labelled with microspheres synthesized by the emulsion of the ionizing radiation technique. The labelling of the cells by means of microspheres containing an iron core produced successful separation of B from T lymphocytes by means of a magnetic field.

Rembaum, A.↗

NASA NTRS1979

Laboratory simulations of the Viking Labeled Release experiment - Kinetics following second nutrient injection and the nature of the gaseous end product

Injection of (C-14)-labeled nutrient onto Mars soil produced an evolution of C-14 gas in the Viking Labeled Release (LR) experiment. However, a second injection of nutrient seven days later was followed by an abrupt diminution of the amount of radioactive gas in the test cell. Simulation experiments performed in the LR Test Standards Module (TSM) have yielded a plausible explanation for this diminution. Radioactive carbon gases were injected into the TSM test cell in the presence and absence of two Mars analog soils. After equilibration, water was injected and its effect observed. The results indicate that the flight data following second nutrient injection can be explained on a physico-chemical basis involving a carbon dioxide/water/soil equilibrium in the test cell. The results also suggest that the gaseous end product of the Labeled Release reaction on Mars is more likely carbon dioxide than carbon monoxide.

Levin, G. V.↗

NASA NTRS1981

Structure of the knowledge base for an expert labeling system

One of the principal objectives of the NASA AgRISTARS program is the inventory of global crop resources using remotely sensed data gathered by Land Satellites (LANDSAT). A central problem in any such crop inventory procedure is the interpretation of LANDSAT images and identification of parts of each image which are covered by a particular crop of interest. This task of labeling is largely a manual one done by trained human analysts and consequently presents obstacles to the development of totally automated crop inventory systems. However, development in knowledge engineering as well as widespread availability of inexpensive hardware and software for artificial intelligence work offers possibilities for developing expert systems for labeling of crops. Such a knowledge based approach to labeling is presented.

Rajaram, N. S.↗

NASA NTRS1981

Development and evaluation of an automatic labeling technique for spring small grains

A labeling technique is described which seeks to associate a sampling entity with a particular crop or crop group based on similarity of growing season and temporal-spectral patterns of development. Human analyst provide contextual information, after which labeling decisions are made automatically. Results of a test of the technique on a large, multi-year data set are reported. Grain labeling accuracies are similar to those achieved by human analysis techniques, while non-grain accuracies are lower. Recommendations for improvments and implications of the test results are discussed.

Crist, E. P.↗

NASA NTRS1980

Transition year labeling error characterization study

Labeling errors made in the large area crop inventory experiment transition year estimates by Earth Observation Division image analysts are identified and quantified. The analysis was made from a subset of blind sites in six U.S. Great Plains states (Oklahoma, Kansas, Montana, Minnesota, North and South Dakota). The image interpretation basically was well done, resulting in a total omission error rate of 24 percent and a commission error rate of 4 percent. The largest amount of error was caused by factors beyond the control of the analysts who were following the interpretation procedures. The odd signatures, the largest error cause group, occurred mostly in areas of moisture abnormality. Multicrop labeling was tabulated showing the distribution of labeling for all crops.

Clinton, N. J.↗

NASA NTRS1981

Evaluation of the US/Canada wheat and barley exploratory experiment shakedown test analyst labeling results

Labeling accuracies using the reformatted labeling procedure are evaluated. The procedure is described and the decision logic applied to six LANDSAT segments from the 1978 crop year. Small grains and nonsmall grains are labeled, and small grains differentiated from barley. An error analysis was performed. Recommended changes to the reformatted procedure are presented.

Carnes, J. G.↗

NASA NTRS1984

Labeling Solar-Cell Modules

Photocopying machine produces durable identification label. Process used for double glass photovoltaic-cell modules. Matrix of cells sandwiched between thin, flexible glass mats and covered above and below by protective sheets of glass. Label contains such information as manufacturer, model number, voltage and power ratings, and serial number. May also contain electrical-shock hazard warning and identification of positive and negative terminals. Method saves expense of procuring and applying conventional labels.

Watson, E. G.↗

NASA NTRS1992

Using partially labeled data for normal mixture identification with application to class definition

The problem of estimating the parameters of a normal mixture density when, in addition to the unlabeled samples, sets of partially labeled samples are available is addressed. The density of the multidimensional feature space is modeled with a normal mixture. It is assumed that the set of components of the mixture can be partitioned into several classes and that training samples are available from each class. Since for any training sample the class of origin is known but the exact component of origin within the corresponding class is unknown, the training samples as considered to be partially labeled. The EM iterative equations are derived for estimating the parameters of the normal mixture in the presence of partially labeled samples. These equations can be used to combine the supervised and nonsupervised learning processes.

Shahshahani, Behzad M.↗