A Fast Cluster-Assumption Based Active-Learning Technique for Classification of Remote Sensing Images.
Swarnajyoti PatraLorenzo BruzzonePublished in: IEEE Trans. Geosci. Remote. Sens. (2011)
Keyphrases
- remote sensing images
- active learning
- multi spectral images
- change detection
- landsat etm
- remote sensing
- multispectral
- pattern recognition
- spectral features
- supervised learning
- unsupervised classification
- classification accuracy
- machine learning
- feature extraction
- feature selection
- semi supervised
- text classification
- landsat tm
- decision trees
- pattern classification
- multispectral images
- land cover
- remote sensing data
- support vector machine
- support vector
- batch mode active learning
- computer vision
- remotely sensed images
- training set
- selective sampling
- supervised classification
- learning process
- hyperspectral images
- rare class
- active learning framework
- satellite images
- data points
- unsupervised learning