A Trace Lasso Regularized L1-norm Graph Cut for Highly Correlated Noisy Hyperspectral Image.
Ramanarayan MohantyS. L. HappyNilesh SutharAurobinda RoutrayPublished in: CoRR (2018)
Keyphrases
- graph cuts
- highly correlated
- hyperspectral images
- hyperspectral
- least squares
- energy function
- markov random field
- average precision
- remote sensing
- multispectral
- image segmentation
- statistically significant
- mean shift
- graph cut segmentation
- feature selection
- high dimensionality
- segmentation algorithm
- sparse representation
- multispectral images
- higher order
- data sets
- high dimensional data
- support vector
- high quality
- target detection