Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “unsupervised”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
NASA NTRS2003

A Fast Implementation of the ISOCLUS Algorithm

Unsupervised clustering is a fundamental building block in numerous image processing applications. One of the most popular and widely used clustering schemes for remote sensing applications is the ISOCLUS algorithm, which is based on the ISODATA method. The algorithm is given a set of n data points in d-dimensional space, an integer k indicating the initial number of clusters, and a number of additional parameters. The general goal is to compute the coordinates of a set of cluster centers in d-space, such that those centers minimize the mean squared distance from each data point to its nearest center. This clustering algorithm is similar to another well-known clustering method, called k-means. One significant feature of ISOCLUS over k-means is that the actual number of clusters reported might be fewer or more than the number supplied as part of the input. The algorithm uses different heuristics to determine whether to merge lor split clusters. As ISOCLUS can run very slowly, particularly on large data sets, there has been a growing .interest in the remote sensing community in computing it efficiently. We have developed a faster implementation of the ISOCLUS algorithm. Our improvement is based on a recent acceleration to the k-means algorithm of Kanungo, et al. They showed that, by using a kd-tree data structure for storing the data, it is possible to reduce the running time of k-means. We have adapted this method for the ISOCLUS algorithm, and we show that it is possible to achieve essentially the same results as ISOCLUS on large data sets, but with significantly lower running times. This adaptation involves computing a number of cluster statistics that are needed for ISOCLUS but not for k-means. Both the k-means and ISOCLUS algorithms are based on iterative schemes, in which nearest neighbors are calculated until some convergence criterion is satisfied. Each iteration requires that the nearest center for each data point be computed. Naively, this requires O(kn) time, where k denotes the current number of centers. Traditional techniques for accelerating nearest neighbor searching involve storing the k centers in a data structure. However, because of the iterative nature of the algorithm, this data structure would need to be rebuilt with each new iteration. Our approach is to store the data points in a kd-tree data structure. The assignment of points to nearest neighbors is carried out by a filtering process, which successively eliminates centers that can not possibly be the nearest neighbor for a given region of space. This algorithm is significantly faster, because large groups of data points can be assigned to their nearest center in a single operation. Preliminary results on a number of real Landsat datasets show that our revised ISOCLUS-like scheme runs about twice as fast.

Memarsadeghi, Nargess↗
NASA NTRS2003

How Sample Completeness Affects Gamma-Ray Burst Classification

Unsupervised pattern-recognition algorithms support the existence of three gamma-ray burst classes: class 1 (long, large-fluence bursts of intermediate spectral hardness), class 2 (short, small-fluence, hard bursts), and class 3 (soft bursts of intermediate durations and fluences). The algorithms surprisingly assign larger membership to class 3 than to either of the other two classes. A known systematic bias has been previously used to explain the existence of class 3 in terms of class 1 ; this bias allows the fluences and durations of some bursts to be underestimated, as recently shown by Hakkila et al. We show that this bias primarily affects only the longest bursts and cannot explain the bulk of the class 3 properties. We resolve the question of class 3's existence by demonstrating how samples obtained using standard trigger mechanisms fail to preserve the duration characteristics of small-peak flux bursts. Sample incompleteness is thus primarily responsible for the existence of class 3. In order to avoid this incompleteness, we show how a new, dual-timescale peak flux can be defined in terms of peak flux and fluence. The dual-timescale peak flux preserves the duration distribution of faint bursts and correlates better with spectral hardness (and presumably redshift) than either peak flux or fluence. The techniques presented here are generic and have applicability to the studies of other transient events. The results also indicate that pattern recognition algorithms are sensitive to sample completeness; this can influence the study of large astronomical databases, such as those found in a virtual observatory.

Hakkila, Jon↗
NASA NTRS2002

Low Altitude AVIRIS Data for Mapping Landform Types on West Ship Island, Mississippi

A chain of barrier islands provides protection against hurricanes and severe storms along the south and southeastern shores of the United States. Barrier island landform types can be spectrally similar and as small as a few meters across, making highly detailed maps difficult to produce. To determine whether high-resolution airborne hyperspectral imagery could provide detailed maps of barrier island landform types, we used low-altitude hyperspectral and multispectral imagery to map surface environments of West Ship Island, Mississippi. We employed 3.4-meter AVIRIS hyperspectral imagery acquired in July 1999 and 0.5-meter ADAR multispectral data acquired in November 1997. The data were co-registered to digital ortho aerial imagery, and the AVIRIS data was scaled to ground reflectance using ATREM software. Unsupervised classification of AVIRIS and ADAR data proceeded using ISODATA clustering techniques. The resulting landform maps were field-checked and compared to aerial photography and digital elevation maps. Preliminary analyses indicated that the AVIRIS classification mapped more landform types, while the ADAR-based map enabled smaller patches to be identified. Used together, these maps provided a means to assess landform distributions of West Ship Island before and after Hurricane Gorges. Classification accuracy is being addressed through photo-interpretation and field surveys of sample areas selected with stratified random sampling.

Spruce, Joseph↗
NASA NTRS2003

Low Altitude AVIRIS Data for Mapping Landform Types on West Ship Island, Mississippi

A chain of barrier islands provides protection against hurricanes and severe storms along the southern and southeastern shores of the Unites States. Barrier island landform types can be spectrally similar and as small as a few meters across, making highly detailed maps difficult to produce. To determine whether high-resolution airborne hyperspectral imagery could provide detailed maps of barrier island landform types, we used low-altitude hyperspectral and multispectral imagery to map surface environments of West Ship Island, Mississippi. We employed 3.4 meter AVIRIS hyperspectral imagery acquired in July 1999 and 0.5 meter ADAR multispectral data acquired in November 1997. The data were co-registered to digital ortho aerial imagery, and the AVIRIS data was scaled to ground reflectance using ATREM software. Unsupervised classification of AVIRIS and ADAR data proceeded using ISODATA clustering techniques. The resulting landform maps were field-checked and compared to aerial photography and digital elevation maps. Preliminary analyses indicated that the AVIRIS classification mapped more landform types, while the ADAR-based map enabled smaller patches to be identified. Used together, these maps provided a means to assess landform distributions of West Ship Island before and after Hurricane Georges. Classification accuracy is being assessed through photo-interpretation and field surveys of sample areas selected with stratified random sampling.

Spruce, Joseph↗
NASA NTRS2001

An Approach to Application Validation of Multispectral Sensors Using AVIRIS

High-resolution multispectral data are becoming widely available for commercial and scientific use. For specific applications, such as agriculture studies, there is a need to quantify the performance of such systems. In many cases, parameters such as GSD and SNR can be optimized. Data sets with varying GSD's for the Landsat ETM+ bands were produced to evaluate the effects of GSD on various algorithms and transformations, such as NDVI, principal component analysis, unsupervised classification, and mixture analysis. By showing that AVIRIS data can be used to simulate spaceborne and airborne multispectral platforms over a wide range of GSD, this research can be used to assist in band selection and spatial resolution specifications for new sensors and in optimization of acquisition strategies for existing multispectral systems.

Warner, Amanda↗
NASA NTRS2004

Fossil Signatures Using Elemental Abundance Distributions and Bayesian Probabilistic Classification

Elemental abundances (C6, N7, O8, Na11, Mg12, Al3, P15, S16, Cl17, K19, Ca20, Ti22, Mn25, Fe26, and Ni28) were obtained for a set of terrestrial fossils and the rock matrix surrounding them. Principal Component Analysis extracted five factors accounting for the 92.5% of the data variance, i.e. information content, of the elemental abundance data. Hierarchical Cluster Analysis provided unsupervised sample classification distinguishing fossil from matrix samples on the basis of either raw abundances or PCA input that agreed strongly with visual classification. A stochastic, non-linear Artificial Neural Network produced a Bayesian probability of correct sample classification. The results provide a quantitative probabilistic methodology for discriminating terrestrial fossils from the surrounding rock matrix using chemical information. To demonstrate the applicability of these techniques to the assessment of meteoritic samples or in situ extraterrestrial exploration, we present preliminary data on samples of the Orgueil meteorite. In both systems an elemental signature produces target classification decisions remarkably consistent with morphological classification by a human expert using only structural (visual) information. We discuss the possibility of implementing a complexity analysis metric capable of automating certain image analysis and pattern recognition abilities of the human eye using low magnification optical microscopy images and discuss the extension of this technique across multiple scales.

Hoover, Richard B.↗
NASA NTRS2005

Techniques for Producing Coastal Land Water Masks from Landsat and Other Multispectral Satellite Data

Coastal erosion and land loss continue to threaten many areas in the United States. Landsat data has been used to monitor regional coastal change since the 1970s. Many techniques can be used to produce coastal land water masks, including image classification and density slicing of individual bands or of band ratios. Band ratios used in land water detection include several variations of the Normalized Difference Water Index (NDWI). This poster discusses a study that compares land water masks computed from unsupervised Landsat image classification with masks from density-sliced band ratios and from the Landsat TM band 5. The greater New Orleans area is employed in this study, due to its abundance of coastal habitats and its vulnerability to coastal land loss. Image classification produced the best results based on visual comparison to higher resolution satellite and aerial image displays. However, density sliced NDWI imagery from either near infrared (NIR) and blue bands or from NIR and green bands also produced more effective land water masks than imagery from the density-sliced Landsat TM band 5. NDWI based on NIR and green bands is noteworthy because it allows land water masks to be generated from multispectral satellite sensors without a blue band (e.g., ASTER and Landsat MSS). NDWI techniques also have potential for producing land water masks from coarser scaled satellite data, such as MODIS.

Spruce, Joseph P.↗
NASA NTRS2005

Searching for 'Unknown Unknowns'

The NASA Engineering and Safety Center (NESC) was established to improve safety through engineering excellence within NASA programs and projects. As part of this goal, methods are being investigated to enable the NESC to become proactive in identifying areas that may be precursors to future problems. The goal is to find unknown indicators of future problems, not to duplicate the program-specific trending efforts. The data that is critical for detecting these indicators exist in a plethora of dissimilar non-conformance and other databases (without a common format or taxonomy). In fact, much of the data is unstructured text. However, one common database is not required if the right standards and electronic tools are employed. Electronic data mining is a particularly promising tool for this effort into unsupervised learning of common factors. This work in progress began with a systematic evaluation of available data mining software packages, based on documented decision techniques using weighted criteria. The four packages, which were perceived to have the most promise for NASA applications, are being benchmarked and evaluated by independent contractors. Preliminary recommendations for "best practices" in data mining and trending are provided. Final results and recommendations should be available in the Fall 2005. This critical first step in identifying "unknown unknowns" before they become problems is applicable to any set of engineering or programmatic data.

Parsons, Vickie S.↗
NASA NTRS2006

Enabling the Discovery of Recurring Anomalies in Aerospace System Problem Reports using High-Dimensional Clustering Techniques

This paper describes the results of a significant research and development effort conducted at NASA Ames Research Center to develop new text mining techniques to discover anomalies in free-text reports regarding system health and safety of two aerospace systems. We discuss two problems of significant importance in the aviation industry. The first problem is that of automatic anomaly discovery about an aerospace system through the analysis of tens of thousands of free-text problem reports that are written about the system. The second problem that we address is that of automatic discovery of recurring anomalies, i.e., anomalies that may be described m different ways by different authors, at varying times and under varying conditions, but that are truly about the same part of the system. The intent of recurring anomaly identification is to determine project or system weakness or high-risk issues. The discovery of recurring anomalies is a key goal in building safe, reliable, and cost-effective aerospace systems. We address the anomaly discovery problem on thousands of free-text reports using two strategies: (1) as an unsupervised learning problem where an algorithm takes free-text reports as input and automatically groups them into different bins, where each bin corresponds to a different unknown anomaly category; and (2) as a supervised learning problem where the algorithm classifies the free-text reports into one of a number of known anomaly categories. We then discuss the application of these methods to the problem of discovering recurring anomalies. In fact the special nature of recurring anomalies (very small cluster sizes) requires incorporating new methods and measures to enhance the original approach for anomaly detection. ?& pant 0-

Srivastava, Ashok, N.↗
NASA NTRS1995

Introduction to Concepts in Artificial Neural Networks

This introduction to artificial neural networks summarizes some basic concepts of computational neuroscience and the resulting models of artificial neurons. The terminology of biological and artificial neurons, biological and machine learning and neural processing is introduced. The concepts of supervised and unsupervised learning are explained with examples from the power system area. Finally, a taxonomy of different types of neurons and different classes of artificial neural networks is presented.

Artificial Neural↗
NASA NTRS2006

Anomaly Detection in Large Sets of High-Dimensional Symbol Sequences

This paper addresses the problem of detecting and describing anomalies in large sets of high-dimensional symbol sequences. The approach taken uses unsupervised clustering of sequences using the normalized longest common subsequence (LCS) as a similarity measure, followed by detailed analysis of outliers to detect anomalies. As the LCS measure is expensive to compute, the first part of the paper discusses existing algorithms, such as the Hunt-Szymanski algorithm, that have low time-complexity. We then discuss why these algorithms often do not work well in practice and present a new hybrid algorithm for computing the LCS that, in our tests, outperforms the Hunt-Szymanski algorithm by a factor of five. The second part of the paper presents new algorithms for outlier analysis that provide comprehensible indicators as to why a particular sequence was deemed to be an outlier. The algorithms provide a coherent description to an analyst of the anomalies in the sequence, compared to more normal sequences. The algorithms we present are general and domain-independent, so we discuss applications in related areas such as anomaly detection.

Budalakoti, Suratna↗
NASA NTRS2006

Automated Classification of Thermal Infrared Spectra Using Self-organizing Maps

Existing and planned space missions to a variety of planetary and satellite surfaces produce an ever increasing volume of spectral data. Understanding the scientific informational content in this large data volume is a daunting task. Fortunately various statistical approaches are available to assess such data sets. Here we discuss an automated classification scheme based on Kohonen Self-organizing maps (SOM) we have developed. The SUM process produces an output layer were spectra having similar properties lie in close proximity to each other. One major effort is partitioning this output layer into appropriate regions. This is prefonned by defining dosed regions based upon the strength of the boundaries between adjacent cells in the SOM output layer. We use the Davies-Bouldin index as a measure of the inter-class similarities and intra-class dissimilarities that determines the optimum partition of the output layer, and hence number of SOM clusters. This allows us to identify the natural number of clusters formed from the spectral data. Mineral spectral libraries prepared at Arizona State University (ASU) and John Hopkins University (JHU) are used to test and evaluate the classification scheme. We label the library sample spectra in a hierarchical scheme with class, subclass, and mineral group names. We use a portion of the spectra to train the SOM, i.e. produce the output layer, while the remaining spectra are used to test the SOM. The test spectra are presented to the SOM output layer and assigned membership to the appropriate cluster. We then evaluate these assignments to assess the scientific meaning and accuracy of the derived SOM classes as they relate to the labels. We demonstrate that unsupervised classification by SOMs can be a useful component in autonomous systems designed to identify mineral species from reflectance and emissivity spectra in the therrnal IR.

Roush, Ted L.↗
NASA NTRS2005

Techniques for Producing Coastal Land Water Masks from Landsat and Other Multispectral Satellite Data

Coastal erosion and land loss continue to threaten many areas in the United States. Landsat data has been used to monitor regional coastal change since the 1970's. Many techniques can be used to produce coastal land water masks, including image classification and density slicing of individual bands or of band ratios. Band ratios used in land water detection include several variations of the Normalized Difference Water Index (NDWI). This poster discusses a study that compares land water masks computed from unsupervised Landsat image classification with masks from density-sliced band ratios and from the Landsat TM band 5. The greater New Orleans area is imployed in this study, due to its abundance of coastal habitats and ist vulnerability to coastal land loss. Image classification produced the best results based on visual comparison to higher resolution satellite and aerial image displays. However, density-sliced NDWI imagery from either near infrared (NIR) and blue bands or from NIR and green bands also produced more effective land water masks than imagery from the density-sliced Landsat TM band 5. NDWI based on NIR and green bands is noteworthy because it allows land water masks to be generated form multispectral satellite sensors without a blue band (e.g., ASTER and Landsat MSS). NDWI techniques also have potential for producing land water masks from coarser scaled satellite data, such as MODIS.

Spruce, Joe↗
NASA NTRS2007

Morphological Feature Extraction for Automatic Registration of Multispectral Images

The task of image registration can be divided into two major components, i.e., the extraction of control points or features from images, and the search among the extracted features for the matching pairs that represent the same feature in the images to be matched. Manual extraction of control features can be subjective and extremely time consuming, and often results in few usable points. On the other hand, automated feature extraction allows using invariant target features such as edges, corners, and line intersections as relevant landmarks for registration purposes. In this paper, we present an extension of a recently developed morphological approach for automatic extraction of landmark chips and corresponding windows in a fully unsupervised manner for the registration of multispectral images. Once a set of chip-window pairs is obtained, a (hierarchical) robust feature matching procedure, based on a multiresolution overcomplete wavelet decomposition scheme, is used for registration purposes. The proposed method is validated on a pair of remotely sensed scenes acquired by the Advanced Land Imager (ALI) multispectral instrument and the Hyperion hyperspectral instrument aboard NASA's Earth Observing-1 satellite.

Plaza, Antonio↗
NASA NTRS2006

Data Mining SIAM Presentation

This viewgraph document describes the data mining system developed at NASA Ames. Many NASA programs have large numbers (and types) of problem reports.These free text reports are written by a number of different people, thus the emphasis and wording vary considerably With so much data to sift through, analysts (subject experts) need help identifying any possible safety issues or concerns and help them confirm that they haven't missed important problems. Unsupervised clustering is the initial step to accomplish this; We think we can go much farther, specifically, identify possible recurring anomalies. Recurring anomalies may be indicators of larger systemic problems. The requirement to identify these anomalies has led to the development of Recurring Anomaly Discovery System (ReADS).

Srivastava, Ashok↗
NASA NTRS2008

An Investigation of State-Space Model Fidelity for SSME Data

In previous studies, a variety of unsupervised anomaly detection techniques for anomaly detection were applied to SSME (Space Shuttle Main Engine) data. The observed results indicated that the identification of certain anomalies were specific to the algorithmic method under consideration. This is the reason why one of the follow-on goals of these previous investigations was to build an architecture to support the best capabilities of all algorithms. We appeal to that goal here by investigating a cascade, serial architecture for the best performing and most suitable candidates from previous studies. As a precursor to a formal ROC (Receiver Operating Characteristic) curve analysis for validation of resulting anomaly detection algorithms, our primary focus here is to investigate the model fidelity as measured by variants of the AIC (Akaike Information Criterion) for state-space based models. We show that placing constraints on a state-space model during or after the training of the model introduces a modest level of suboptimality. Furthermore, we compare the fidelity of all candidate models including those embodying the cascade, serial architecture. We make recommendations on the most suitable candidates for application to subsequent anomaly detection studies as measured by AIC-based criteria.

Martin, Rodney Alexander↗