A spatial-spectral kernel-based approach for the classification of remote-sensing images.
Mathieu FauvelJocelyn ChanussotJón Atli BenediktssonPublished in: Pattern Recognit. (2012)
Keyphrases
- remote sensing images
- spectral features
- multi spectral images
- multispectral
- remote sensing
- change detection
- landsat etm
- multispectral images
- remote sensing data
- spectral characteristics
- support vector
- hyperspectral
- hyperspectral data
- pattern recognition
- feature space
- hyperspectral images
- geographical information systems
- landsat tm
- remotely sensed images
- unsupervised classification
- remote sensing imagery
- land cover
- spatial data
- feature selection
- feature vectors
- spatial databases
- pattern classification
- kernel function
- data sets
- satellite images
- supervised learning
- human skin
- image classification
- marked point processes
- spectral analysis
- hyperspectral imagery
- feature extraction
- color images
- computer vision
- machine learning