A novel approach for semi-supervised classification of remote sensing images using a clustering-based selection of training data according to their GMM responsibilities.
Victor-Emil NeagoeVlad Chirila-BerbenteaPublished in: IGARSS (2017)
Keyphrases
- remote sensing images
- semi supervised classification
- training data
- unlabeled data
- labeled data
- semi supervised learning
- change detection
- remote sensing
- multispectral
- semi supervised
- supervised learning
- marked point processes
- co training
- data sets
- satellite images
- training set
- training examples
- test data
- hyperspectral
- training process
- landsat etm
- active learning
- prior knowledge
- learning algorithm
- transfer learning
- feature vectors
- pairwise
- decision trees
- supervised classification
- training samples
- labeled and unlabeled data
- image data
- reinforcement learning
- neural network