Improvement of hyperspectral classification accuracy with limited training data using meanshift segmentation.
Okan Bilge OzdemirYasemin Yardimci ÇetinPublished in: SIU (2014)
Keyphrases
- hyperspectral
- classification accuracy
- training data
- pixel classification
- mean shift
- remote sensing
- hyperspectral images
- multispectral
- color image segmentation
- training set
- infrared
- hyperspectral imagery
- feature space
- naive bayes
- hyperspectral data
- image data
- segmentation algorithm
- hyperspectral imaging
- multi band
- hyperspectral remote sensing
- segmented images
- hyperspectral image classification
- target detection
- learning algorithm
- image segmentation
- image analysis
- information content
- feature selection
- decision trees
- particle filter
- segmentation method
- spatial resolution
- satellite images
- support vector
- region growing
- spectral resolution
- reflectance spectra
- graph cuts
- frequency domain
- support vector machine
- spectral bands