NASA Jet Propulsion Laboratory California Institute of Technology Follow this link to skip to the main content

BEACON eSpace at Jet Propulsion Laboratory >
JPL Technical Report Server >
JPL TRS 1992+ >

Please use this identifier to cite or link to this item: http://hdl.handle.net/2014/39811

Title: Onboard classifiers for science event detection on a remote sensing spacecraft
Authors: Castano, Rebecca
Mazzoni, Dominic
Tang, Nghia
Greeley, Ron
Doggett, Thomas
Cichy, Ben
Chien, Steve
Davies, Ashley
Keywords: feature detection
Support Vector Machine (SVM)
cyrosphere
machine learning
classification
Issue Date: 20-Aug-2006
Publisher: Pasadena, CA : Jet Propulsion Laboratory, National Aeronautics and Space Administration, 2006.
Citation: Knowledge Discovery and Data Mining, Philadelphia, Pennsylvania, August 20-23, 2006.
Abstract: Typically, data collected by a spacecraft is downlinked to Earth and pre-processed before any analysis is performed. We have developed classifiers that can be used onboard a spacecraft to identify high priority data for downlink to Earth, providing a method for maximizing the use of a potentially bandwidth limited downlink channel. Onboard analysis can also enable rapid reaction to dynamic events, such as flooding, volcanic eruptions or sea ice break-up. Four classifiers were developed to identify cryosphere events using hyperspectral images. These classifiers include a manually constructed classifier, a Support Vector Machine (SVM), a Decision Tree and a classifier derived by searching over combinations of thresholded band ratios. Each of the classifiers was designed to run in the computationally constrained operating environment of the spacecraft. A set of scenes was hand-labeled to provide training and testing data. Performance results on the test data indicate that the SVM and manual classifiers outperformed the Decision Tree and band-ratio classifiers with the SVM yielding slightly better classifications than the manual classifier.
URI: http://hdl.handle.net/2014/39811
Appears in Collections:JPL TRS 1992+

Files in This Item:

File Description SizeFormat
06-1687.pdf198.47 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, but are furnished with U.S. government purpose use rights.

 

Privacy/Copyright Image Policy Beacon Home Contact Us
NASA Home Page + Div 27
+ JPL Space
Site last updated on November 15, 2012.
If you have any comments or suggestions for this web site, please e-mail Alexander Smith or call 4-4202.