Ensemble Margin Based Semi-Supervised Random Forest for the Classification of Hyperspectral Image with Limited Training Data.
Wei FengWenjiang HuangGabriel DauphinJunshi XiaYinghui QuanHuichun YeYingying DongPublished in: IGARSS (2019)
Keyphrases
- random forest
- decision trees
- training data
- hyperspectral images
- feature set
- random forests
- semi supervised
- supervised learning
- unlabeled data
- classification accuracy
- ensemble classifier
- training set
- ensemble methods
- hyperspectral data
- class labels
- base classifiers
- labeled data
- decision tree learning algorithms
- classification models
- feature selection
- hyperspectral
- ensemble learning
- semi supervised learning
- multi class
- support vector machine
- training samples
- machine learning algorithms
- naive bayes
- data sets
- hyperspectral image classification
- feature extraction
- multi label
- machine learning methods
- image classification
- machine learning
- learning algorithm
- test data
- benchmark datasets
- pattern recognition
- generalization error
- feature space
- multispectral
- remote sensing
- svm classifier
- test set
- pairwise
- feature vectors
- active learning
- unsupervised learning
- support vector
- base learners