Information Fusion in Kernel-Induced Spaces for Robust Subpixel Hyperspectral ATR

Hyperspectral-based automatic target recognition (ATR) and classification systems often project the high-dimensional hyperspectral reflectance signatures onto a lower dimensional subspace using techniques such as principal component analysis, Fisher's linear discriminant analysis (LDA), and ste...

Full description

Saved in:
Bibliographic Details
Published in:IEEE geoscience and remote sensing letters 2009-07, Vol.6 (3), p.572-576
Main Authors: Prasad, S., Bruce, L.M.
Format: Article
Language:English
Subjects:
Classification
Decision fusion
Discriminant analysis
Hyperspectral imaging
Hyperspectral sensors
Kernel
kernel methods
Kernels
Linear discriminant analysis
multiclassifiers
Partitions
Pattern classification
Principal component analysis
Projection
Recognition
Reflectivity
Robustness
Studies
Subspaces
Target recognition
Training data
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Hyperspectral-based automatic target recognition (ATR) and classification systems often project the high-dimensional hyperspectral reflectance signatures onto a lower dimensional subspace using techniques such as principal component analysis, Fisher's linear discriminant analysis (LDA), and stepwise LDA. In a general classification framework, these projections are suboptimal and, in the absence of sufficient training data, are likely to be ill conditioned. In recent work, the authors proposed a divide-and-conquer approach that partitions the hyperspectral space into contiguous subspaces followed by a multiclassifier and decision-fusion (MCDF) framework. Although this technique alleviated the small-sample-size problem and provided a good recognition performance in light and moderate pixel mixing, the performance significantly decreased under severe mixing conditions, as it does with conventional ATR techniques. In this letter, the authors propose a kernel discriminant analysis-based projection in each subspace of the partition, followed by the MCDF framework to ensure robust recognition even in severe pixel-mixing conditions. The performance of the proposed system (as measured by overall recognition accuracies) is greatly superior to conventional dimensionality-reduction techniques as well as the more recently proposed LDA-based MCDF technique.
ISSN:1545-598X
1558-0571
DOI:10.1109/LGRS.2009.2022852