Full-Learning Rotational Quaternion Convolutional Neural Networks and Confluence of Differently Represented Data for PolSAR Land Classification.
Yuya MatsumotoRyo NatsuakiAkira HirosePublished in: IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. (2022)
Keyphrases
- prior knowledge
- data sets
- learning algorithm
- training data
- data sources
- learning process
- background knowledge
- unsupervised learning
- supervised learning
- bayesian methods
- incremental learning
- data quality
- missing data
- data mining techniques
- support vector machine
- data points
- active learning
- data analysis
- support vector
- reinforcement learning
- learned models
- learning phase
- neural network
- classification models
- image sequences
- classification method
- spatial data
- machine learning algorithms
- training samples
- image classification
- database