Improving Imbalanced Land Cover Classification with K-Means SMOTE: Detecting and Oversampling Distinctive Minority Spectral Signatures.
João FonsecaGeorgios DouzasFernando BaçãoPublished in: Inf. (2021)
Keyphrases
- minority class
- land cover classification
- class imbalance
- hyperspectral
- class imbalanced
- majority class
- spectral signatures
- class distribution
- k means
- imbalanced data
- multispectral
- imbalanced data sets
- classification error
- imbalanced datasets
- cost sensitive learning
- nearest neighbour
- support vector machine
- decision boundary
- supervised classification
- clustering algorithm
- remote sensing
- remote sensing data
- hyperspectral imagery
- hyperspectral images
- training set
- original data
- cost sensitive
- land cover
- sampling methods
- spatial resolution
- training dataset
- training data
- hyperspectral data
- computer vision
- data sets
- target detection
- remote sensing images
- information content
- class labels
- frequency domain
- image data
- data streams
- image segmentation
- image processing