NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Metric Learning for Hyperspectral Image SegmentationWe present a metric learning approach to improve the performance of unsupervised hyperspectral image segmentation. Unsupervised spatial segmentation can assist both user visualization and automatic recognition of surface features. Analysts can use spatially-continuous segments to decrease noise levels and/or localize feature boundaries. However, existing segmentation methods use tasks-agnostic measures of similarity. Here we learn task-specific similarity measures from training data, improving segment fidelity to classes of interest. Multiclass Linear Discriminate Analysis produces a linear transform that optimally separates a labeled set of training classes. The defines a distance metric that generalized to a new scenes, enabling graph-based segmentation that emphasizes key spectral features. We describe tests based on data from the Compact Reconnaissance Imaging Spectrometer (CRISM) in which learned metrics improve segment homogeneity with respect to mineralogical classes.
Document ID
20150005766
Document Type
Conference Paper
External Source(s)
Authors
Bue, Brian D. (Rice Univ. Houston, TX, United States)
Thompson, David R. (Jet Propulsion Lab., California Inst. of Tech. Pasadena, CA, United States)
Gilmore, Martha S. (Wesleyan Univ. Middletown, CT, United States)
Castano, Rebecca (Jet Propulsion Lab., California Inst. of Tech. Pasadena, CA, United States)
Date Acquired
April 16, 2015
Publication Date
June 6, 2011
Subject Category
Instrumentation and Photography
Meeting Information
IEEE Workshop on Hyperspectral Image Processing: Evolution in Remote Sensing(Lisbon)
Distribution Limits
Public
Copyright
Other
Keywords
Compact Reconnaissance Imaging Spectrometer (CRISM)
segimentation
metric learning