On the use of small training sets for neural network-based characterization of mixed pixels in remotely sensed hyperspectral images

Authors:
Javier Plaza;Antonio Plaza;Rosa Perez;Pablo Martinez
Affiliations:
Neural Networks and Signal Processing Group (GRNPS), Computer Science Department, University of Extremadura, Avda. de la Universidad s/n, 10.071 Caceres, Spain;Neural Networks and Signal Processing Group (GRNPS), Computer Science Department, University of Extremadura, Avda. de la Universidad s/n, 10.071 Caceres, Spain;Neural Networks and Signal Processing Group (GRNPS), Computer Science Department, University of Extremadura, Avda. de la Universidad s/n, 10.071 Caceres, Spain;Neural Networks and Signal Processing Group (GRNPS), Computer Science Department, University of Extremadura, Avda. de la Universidad s/n, 10.071 Caceres, Spain
Venue:
Pattern Recognition
Year:
2009

Citing 8
Cited 2

Independent component analysis, a new concept?

Signal Processing - Special issue on higher order statistics
Incorporating mixed pixels in the training, allocation and testing stages of supervised classifications

Pattern Recognition Letters - Special issue on non-conventional pattern analysis in remote sensing
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Morphological Image Analysis: Principles and Applications

Morphological Image Analysis: Principles and Applications
Hyperspectral Imaging: Techniques for Spectral Detection and Classification

Hyperspectral Imaging: Techniques for Spectral Detection and Classification
Remote Sensing, Third Edition: Models and Methods for Image Processing

Remote Sensing, Third Edition: Models and Methods for Image Processing
Implementation of real-time constrained linear discriminant analysis to remote sensing image classification

Pattern Recognition
Decision boundary feature extraction for neural networks

IEEE Transactions on Neural Networks

Reorganizing neural network system for two spirals and linear low-density polyethylene copolymer problems

Applied Computational Intelligence and Soft Computing
Hyperion hyperspectral imagery analysis combined with machine learning classifiers for land use/cover mapping

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.01

Visualization

Abstract

In this work, neural network-based models involved in hyperspectral image spectra separation are considered. Focus is on how to select the most highly informative samples for effectively training the neural architecture. This issue is addressed here by several new algorithms for intelligent selection of training samples: (1) a border-training algorithm (BTA) which selects training samples located in the vicinity of the hyperplanes that can optimally separate the classes; (2) a mixed-signature algorithm (MSA) which selects the most spectrally mixed pixels in the hyperspectral data as training samples; and (3) a morphological-erosion algorithm (MEA) which incorporates spatial information (via mathematical morphology concepts) to select spectrally mixed training samples located in spatially homogeneous regions. These algorithms, along with other standard techniques based on orthogonal projections and a simple Maximin-distance algorithm, are used to train a multi-layer perceptron (MLP), selected in this work as a representative neural architecture for spectral mixture analysis. Experimental results are provided using both a database of nonlinear mixed spectra with absolute ground truth and a set of real hyperspectral images, collected at different altitudes by the digital airborne imaging spectrometer (DAIS 7915) and reflective optics system imaging spectrometer (ROSIS) operating simultaneously at multiple spatial resolutions.