Automated generation of semi-labeled training samples for nonlinear neural network-based abundance estimation in hyperspectral data.
Javier PlazaAntonio PlazaRosa M. PérezPablo MartínezPublished in: IGARSS (2005)
Keyphrases
- training samples
- hyperspectral data
- supervised learning
- training data
- training set
- hyperspectral
- hyperspectral images
- hyperspectral imagery
- hyperplane
- learning algorithm
- feature space
- test sample
- multispectral
- remote sensing
- high dimensional
- nearest neighbor classifier
- face images
- random projections
- active learning
- image processing
- spatial resolution
- infrared
- data compression
- principal components
- data sets
- principal component analysis
- high dimensional feature space
- support vector machine
- similarity measure
- image sequences
- positive training examples