Change detection and classification of land cover in multispectral satellite imagery using clustering of sparse approximations (CoSA) over learned feature dictionaries.
Daniela I. MoodySteven P. BrumbyJoel C. RowlandGarrett L. AltmannAmy E. LarsonPublished in: AIPR (2014)
Keyphrases
- land cover
- satellite imagery
- multispectral
- change detection
- remote sensing
- land cover classification
- remotely sensed
- remote sensing images
- supervised classification
- sparse approximations
- satellite images
- remotely sensed data
- remote sensing imagery
- land cover change
- remote sensing data
- landsat tm
- remotely sensed images
- hyperspectral data
- hyperspectral images
- hyperspectral
- image analysis
- feature vectors
- spatial resolution
- unsupervised learning
- image data
- normalized difference vegetation index
- sparse representation
- pattern recognition
- high spatial resolution
- multispectral images
- image processing
- hyperspectral imagery
- feature extraction
- supervised learning
- high resolution
- image classification
- digital elevation models
- face recognition
- high quality
- geographic information systems
- data streams
- video sequences
- feature space
- image features
- image registration
- computer vision
- neural network