Search NASA⌕ Search

Engineering topics

Wagstaff, Kiri L.

Publications and source records attributed to Wagstaff, Kiri L..

47 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS2006

Tracking the Martian CO2 Polar Ice Caps in Infrared Images

Researchers at NASA s Jet Propulsion Laboratory have developed a method for automatically tracking the polar caps on Mars as they advance and recede each year (see figure). The seasonal Mars polar caps are composed mainly of CO2 ice and are therefore cold enough to stand out clearly in infrared data collected by the Thermal Emission Imaging System (THEMIS) onboard the Mars Odyssey spacecraft. The Bimodal Image Temperature (BIT) histogram analysis algorithm analyzes raw, uncalibrated data to identify images that contain both "cold" ("polar cap") and "warm" ("not polar cap") pixels. The algorithm dynamically identifies the temperature that separates these two regions. This flexibility is critical, because in the absence of any calibration, the threshold temperature can vary significantly from image to image. Using the identified threshold, the algorithm classifies each pixel in the image as "polar cap" or "not polar cap," then identifies the image row that contains the spatial transition from "polar cap" to "not polar cap." While this method is useful for analyzing data that has already been returned by THEMIS, it has even more significance with respect to data that has not yet been collected. Instead of seeking the polar cap only in specific, targeted images, the simplicity and efficiency of this method makes it feasible for direct, onboard use. That is, THEMIS could continuously monitor its observations for any detections of the polar-cap edge, producing detections over a wide range of spatial and temporal conditions. This effort can greatly contribute to our understanding of long-term climatic change on Mars.

Wagstaff, Kiri L.↗
NASA NTRS2006

When is Constrained Clustering Beneficial, and Why?

Several researchers have shown that constraints can improve the results of a variety of clustering algorithms. However, there can be a large variation in this improvement, even for a fixed number of constraints for a given data set. We present the first attempt to provide insight into this phenomenon by characterizing two constraint set properties: informativeness and coherence. We show that these measures can help explain why some constraint sets are more beneficial to clustering algorithms than others. Since they can be computed prior to clustering, these measures can aid in deciding which constraints to use in practice.

clustering algorithms↗
NASA NTRS2006

Learning User Preferences for Sets of Objects

Most work on preference learning has focused on pairwise preferences or rankings over individual items. In this paper, we present a method for learning preferences over sets of items. Our learning method takes as input a collection of positive examples--that is, one or more sets that have been identified by a user as desirable. Kernel density estimation is used to estimate the value function for individual items, and the desired set diversity is estimated from the average set diversity observed in the collection. Since this is a new learning problem, we introduce a new evaluation methodology and evaluate the learning method on two data collections: synthetic blocks-world data and a new real-world music data collection that we have gathered.

preferences↗
NASA NTRS2005

Towards Onboard Orbital Tracking of Seasonal Polar Volatiles on Mars

Current conditions on Mars support both a residual polar cap, composed mainly of water ice, and a seasonal cap, composed of CO2, which appears and disappears each winter. Kieffer and Titus characterized the recession of the seasonal south polar cap using an arctangent curve fit based on data from the Thermal Emission Spectrometer on Mars Global Surveyor [1]. They also found significant interannual deviations, at the regional scale, in the recession rate [2]. Further observations will enable the refinement of our models of polar cap evolution in both hemispheres. We have developed the Bimodal Image Temperature (BIT) Histogram Analysis method for the automated detection and tracking of the seasonal polar ice caps on Mars. It is specifically tailored for possible use onboard a spacecraft. We have evaluated BIT on uncalibrated data collected by the Thermal Emission Imaging System (THEMIS) instrument [3] on the Mars Odyssey spacecraft. In this paper, we focus on the northern seasonal cap, but our approach is directly applicable to the future analysis of the southern seasonal ice cap as well.

Wagstaff, Kiri L.↗
NASA NTRS2004

Making the most of missing values : object clustering with partial data in astronomy

We demonstrate a clustering analysis algorithm, KSC, that a) uses all observed values and b) does not discard the partially observed objects. KSC uses soft constraints defined by the fully observed objects to assist in the grouping of objects with missing values. We present an analysis of objects taken from the Sloan Digital Sky Survey to demonstrate how imputing the values can be misleading and why the KSC approach can produce more appropriate results.

clustering↗