Active learning paradigms for CBIR systems based on optimum-path forest classification.
André Tavares da SilvaAlexandre X. FalcãoLéo Pini MagalhãesPublished in: Pattern Recognit. (2011)
Keyphrases
- optimum path forest
- learning paradigms
- decision boundary
- nearest neighbor classifier
- image retrieval
- soft computing
- cbir systems
- classification accuracy
- pattern recognition
- relevance feedback
- nearest neighbor
- supervised learning
- image classification
- pattern classification
- bayesian classifier
- classification models
- unsupervised learning
- low level features
- training samples
- machine learning
- image database
- feature space
- image segmentation
- feature selection