Extraction of a Specific Land-Cover Class from Very High Spatial Resolution Imagery Using Positive and Unlabeled Learning with Convolutional Neural Networks.
Khelifa DjerririMoussa Sofiane KarouiRéda AdjoudjPublished in: IGARSS (2019)
Keyphrases
- satellite imagery
- remote sensing
- land cover
- convolutional neural networks
- supervised learning
- active learning
- unsupervised learning
- high spatial resolution
- positive and unlabeled examples
- multispectral
- learning algorithm
- land cover classification
- change detection
- satellite images
- supervised classification
- remote sensing images
- remote sensing data
- spatial resolution
- principal component analysis
- reinforcement learning
- multiscale